Commit | Line | Data |
---|---|---|
059ec3d9 | 1 | #!PERL_COMMAND -w |
d5692f86 | 2 | # $Cambridge: exim/src/src/eximstats.src,v 1.7 2005/06/03 14:28:50 steve Exp $ |
059ec3d9 PH |
3 | |
4 | # Copyright (c) 2001 University of Cambridge. | |
5 | # See the file NOTICE for conditions of use and distribution. | |
6 | ||
7 | # Perl script to generate statistics from one or more Exim log files. | |
8 | ||
9 | # Usage: eximstats [<options>] <log file> <log file> ... | |
10 | ||
11 | # 1996-05-21: Ignore lines not starting with valid date/time, just in case | |
12 | # these get into a log file. | |
13 | # 1996-11-19: Add the -h option to control the size of the histogram, | |
14 | # and optionally turn it off. | |
15 | # Use some Perl 5 things; it should be everywhere by now. | |
16 | # Add the Perl -w option and rewrite so no warnings are given. | |
17 | # Add the -t option to control the length of the "top" listing. | |
18 | # Add the -ne, -nt options to turn off errors and transport | |
19 | # information. | |
20 | # Add information about length of time on queue, and -q<list> to | |
21 | # control the intervals and turn it off. | |
22 | # Add count and percentage of delayed messages to the Received | |
23 | # line. | |
24 | # Show total number of errors. | |
25 | # Add count and percentage of messages with errors to Received | |
26 | # line. | |
27 | # Add information about relaying and -nr to suppress it. | |
28 | # 1997-02-03 Merged in some of the things Nigel Metheringham had done: | |
29 | # Re-worded headings | |
30 | # Added received histogram as well as delivered | |
31 | # Added local senders' league table | |
32 | # Added local recipients' league table | |
33 | # 1997-03-10 Fixed typo "destinationss" | |
34 | # Allow for intermediate address between final and original | |
35 | # when testing for relaying | |
36 | # Give better message when no input | |
37 | # 1997-04-24 Fixed bug in layout of error listing that was depending on | |
38 | # text length (output line got repeated). | |
39 | # 1997-05-06 Bug in option decoding when only one option. | |
40 | # Overflow bug when handling very large volumes. | |
41 | # 1997-10-28 Updated to handle revised log format that might show | |
42 | # HELO name as well as host name before IP number | |
43 | # 1998-01-26 Bugs in the function for calculating the number of seconds | |
44 | # since 1970 from a log date | |
45 | # 1998-02-02 Delivery to :blackhole: doesn't have a T= entry in the log | |
46 | # line; cope with this, thereby avoiding undefined problems | |
47 | # Very short log line gave substring error | |
48 | # 1998-02-03 A routed delivery to a local transport may not have <> in the | |
49 | # log line; terminate the address at white space, not < | |
50 | # 1998-09-07 If first line of input was a => line, $thissize was undefined; | |
51 | # ensure it is zero. | |
52 | # 1998-12-21 Adding of $thissize from => line should have been adding $size. | |
53 | # Oops. Should have looked more closely when fixing the previous | |
54 | # bug! | |
55 | # 1999-11-12 Increased the field widths for printed integers; numbers are | |
56 | # bigger than originally envisaged. | |
57 | # 2001-03-21 Converted seconds() routine to use Time::Local, fixing a bug | |
58 | # whereby seconds($timestamp) - id_seconds($id) gave an | |
59 | # incorrect result. | |
60 | # Added POD documentation. | |
61 | # Moved usage instructions into help() subroutine. | |
62 | # Added 'use strict' and declared all global variables. | |
63 | # Added '-html' flag and resultant code. | |
64 | # Added '-cache' flag and resultant code. | |
65 | # Added add_volume() routine and converted all volume variables | |
66 | # to use it, fixing the overflow problems for individual hosts | |
67 | # on large sites. | |
68 | # Converted all volume output to GB/MB/KB as appropriate. | |
69 | # Don't store local user stats if -nfl is specified. | |
70 | # Modifications done by: Steve Campbell (<steve@computurn.com>) | |
71 | # 2001-04-02 Added the -t_remote_users flag. Steve Campbell. | |
72 | # 2001-10-15 Added the -domain flag. Steve Campbell. | |
73 | # 2001-10-16 Accept files on STDIN or on the command line. Steve Campbell. | |
74 | # 2001-10-21 Removed -domain flag and added -bydomain, -byhost, and -byemail. | |
75 | # We now generate our main parsing subroutine as an eval statement | |
76 | # which improves performance dramatically when not all the results | |
77 | # are required. We also cache the last timestamp to time convertion. | |
78 | # | |
79 | # NOTE: 'Top 50 destinations by (message count|volume)' lines are | |
80 | # now 'Top N (host|email|domain) destinations by (message count|volume)' | |
81 | # where N is the topcount. Steve Campbell. | |
82 | # | |
83 | # 2001-10-30 V1.16 Joachim Wieland. | |
84 | # Fixed minor bugs in add_volume() when taking over this version | |
85 | # for use in Exim 4: -w gave uninitialized value warnings in | |
86 | # two situations: for the first addition to a counter, and if | |
87 | # there were never any gigabytes, thereby leaving the $gigs | |
88 | # value unset. | |
89 | # Initialized $last_timestamp to stop a -w uninitialized warning. | |
90 | # Minor layout tweak for grand totals (nitpicking). | |
91 | # Put the IP addresses for relaying stats in [] and separated by | |
92 | # a space from the domain name. | |
93 | # Removed the IPv4-specific address test when picking out addresses | |
94 | # for relaying. Anything inside [] is OK. | |
95 | # | |
96 | # 2002-07-02 Philip Hazel | |
97 | # Fixed "uninitialized variable" message that occurred for relay | |
98 | # messages that arrived from H=[1.2.3.4] hosts (no name shown). | |
99 | # This bug didn't affect the output. | |
100 | # | |
101 | # 2002-04-15 V1.17 Joachim Wieland. | |
102 | # Added -charts, -chartdir. -chartrel options which use | |
103 | # GD::Graph modules to create graphical charts of the statistics. | |
104 | # | |
105 | # 2002-04-15 V1.18 Steve Campbell. | |
106 | # Added a check for $domain to to stop a -w uninitialized warning. | |
107 | # Added -byemaildomain option. | |
108 | # Only print HTML header links to included tables! | |
109 | # | |
110 | # 2002-08-02 V1.19 Steve Campbell. | |
111 | # Changed the debug mode to dump the parser onto STDERR rather | |
112 | # than STDOUT. Documented the -d flag into the help(). | |
113 | # Rejoined the divergent 2002-04-15 and 2002-07-02 releases. | |
114 | # | |
115 | # 2002-08-21 V1.20 Steve Campbell. | |
116 | # Added the '-merge' option to allow merging of previous reports. | |
117 | # Fixed a missing semicolon when doing -bydomain. | |
118 | # Make volume charts plot the data gigs and bytes rather than just bytes. | |
119 | # Only process log lines with $flag =~ /<=|=>|->|==|\*\*|Co/ | |
120 | # Converted Emaildomain to Edomain - the column header was too wide! | |
121 | # This changes the text output slightly. You can revert to the old | |
122 | # column widths by changing $COLUMN_WIDTHS to 7; | |
123 | # | |
124 | # 2002-09-04 V1.21 Andreas J Mueller | |
125 | # Local deliveries domain now defaults to 'localdomain'. | |
126 | # Don't match F=<From> when looking for the user. | |
127 | # | |
128 | # 2002-09-05 V1.22 Steve Campbell | |
129 | # Fixed a perl 5.005 incompatibility problem ('our' variables). | |
130 | # | |
131 | # 2002-09-11 V1.23 Steve Campbell | |
132 | # Stopped -charts option from throwing errors on null data. | |
133 | # Don't print out 'Errors encountered' unless there are any. | |
134 | ||
135 | # 2002-10-21 V1.23a Philip Hazel - patch from Tony Finch put in until | |
136 | # Steve's eximstats catches up. | |
137 | # Handle log files that include the timezone after the timestamp. | |
138 | # Switch to assuming that log timestamps are in local time, with | |
139 | # an option for UTC timestamps, as in Exim itself. | |
140 | # | |
141 | # 2003-02-05 V1.24 Steve Campbell | |
142 | # Added in Sergey Sholokh's code to convert '<' and '>' characters | |
143 | # in HTML output. Also added code to convert them back with -merge. | |
144 | # Fixed timestamp offsets to convert to seconds rather than minutes. | |
145 | # Updated -merge to work with output files using timezones. | |
146 | # Added cacheing to speed up the calculation of timezone offsets. | |
147 | # | |
148 | # 2003-02-07 V1.25 Steve Campbell | |
149 | # Optimised the usage of mktime() in the seconds subroutine. | |
150 | # Removed the now redundant '-cache' option. | |
151 | # html2txt() now explicitly matches HTML tags. | |
152 | # Implemented a new sorting algorithm - the top_n_sort() routine. | |
153 | # Added Danny Carroll's '-nvr' flag and code. | |
154 | # | |
155 | # 2003-03-13 V1.26 Steve Campbell | |
156 | # Implemented HTML compliance changes recommended by Bernard Massot. | |
157 | # Bug fix to allow top_n_sort() to handle null keys. | |
158 | # Convert all domains and edomains to lowercase. | |
159 | # Remove preceding dots from domains. | |
160 | # | |
161 | # 2003-03-13 V1.27 Steve Campbell | |
162 | # Replaced border attributes with 'border=1', as recommended by | |
163 | # Bernard Massot. | |
164 | # | |
165 | # 2003-06-03 V1.28 John Newman | |
166 | # Added in the ability to skip over the parsing and evaulation of | |
167 | # specific transports as passed to eximstats via the new "-nt/.../" | |
168 | # command line argument. This new switch allows the viewing of | |
169 | # not more accurate statistics but more applicable statistics when | |
170 | # special transports are in use (ie; SpamAssassin). We need to be | |
171 | # able to ignore transports such as this otherwise the resulting | |
172 | # local deliveries are significantly skewed (doubled)... | |
173 | # | |
174 | # 2003-11-06 V1.29 Steve Campbell | |
175 | # Added the '-pattern "Description" "/pattern/"' option. | |
176 | # | |
177 | # 2004-02-17 V1.30 Steve Campbell | |
178 | # Added warnings if required GD::Graph modules are not available or | |
179 | # insufficient -chart* options are specified. | |
180 | # | |
8e669ac1 | 181 | # 2004-02-20 V1.31 Andrea Balzi |
059ec3d9 PH |
182 | # Only show the Local Sender/Destination links if the tables exist. |
183 | # | |
1b4fe9dd PH |
184 | # 2004-07-05 V1.32 Steve Campbell |
185 | # Fix '-merge -h0' divide by zero error. | |
186 | # | |
187 | # 2004-07-15 V1.33 Steve Campbell | |
188 | # Documentation update - I've converted the subroutine | |
189 | # documentation from POD to comments. | |
8974000d SC |
190 | # |
191 | # 2004-12-10 V1.34 Steve Campbell | |
192 | # Eximstats can now parse syslog lines as well as mainlog lines. | |
193 | # | |
884c2af8 | 194 | # 2004-12-20 V1.35 Wouter Verhelst |
d5692f86 | 195 | # Pie charts by volume were actually generated by count. Fixed. |
50adf73a SC |
196 | # |
197 | # 2005-02-07 V1.36 Gregor Herrmann / Steve Campbell | |
198 | # Added average sizes to HTML Top tables. | |
d5692f86 SC |
199 | # |
200 | # 2005-04-26 V1.37 Frank Heydlauf | |
201 | # Added -xls and the ability to specify output files. | |
202 | # | |
203 | # 2005-04-29 V1.38 Steve Campbell | |
204 | # Use FileHandles for outputing results. | |
205 | # Allow any combination of xls, txt, and html output. | |
206 | # Fixed display of large numbers with -nvr option | |
207 | # Fixed merging of reports with empty tables. | |
208 | # | |
209 | # 2005-05-27 V1.39 Steve Campbell | |
210 | # Added the -include_original_destination flag | |
211 | # Removed tabs and trailing whitespace. | |
212 | # | |
213 | # | |
214 | # | |
215 | # For documentation on the logfile format, see | |
216 | # http://www.exim.org/exim-html-4.50/doc/html/spec_48.html#IX2793 | |
059ec3d9 PH |
217 | |
218 | =head1 NAME | |
219 | ||
8974000d | 220 | eximstats - generates statistics from Exim mainlog or syslog files. |
059ec3d9 PH |
221 | |
222 | =head1 SYNOPSIS | |
223 | ||
d5692f86 | 224 | eximstats [Output] [Options] mainlog1 mainlog2 ... |
059ec3d9 PH |
225 | eximstats -merge [Options] report.1.txt report.2.txt ... > weekly_report.txt |
226 | ||
d5692f86 SC |
227 | =head2 Output: |
228 | ||
229 | =over 4 | |
230 | ||
231 | =item B<-txt> | |
232 | ||
233 | Output the results in plain text to STDOUT. | |
234 | ||
235 | =item B<-txt>=I<filename> | |
236 | ||
237 | Output the results in plain text. Filename '-' for STDOUT is accepted. | |
238 | ||
239 | =item B<-html> | |
240 | ||
241 | Output the results in HTML to STDOUT. | |
242 | ||
243 | =item B<-html>=I<filename> | |
244 | ||
245 | Output the results in HTML. Filename '-' for STDOUT is accepted. | |
246 | ||
247 | =item B<-xls> | |
248 | ||
249 | Output the results in Excel compatible Format to STDOUT. | |
250 | Requires the Spreadsheet::WriteExcel CPAN module. | |
251 | ||
252 | =item B<-xls>=I<filename> | |
253 | ||
254 | Output the results in Excel compatible format. Filename '-' for STDOUT is accepted. | |
255 | ||
256 | ||
257 | =back | |
258 | ||
259 | =head2 Options: | |
059ec3d9 PH |
260 | |
261 | =over 4 | |
262 | ||
263 | =item B<-h>I<number> | |
264 | ||
265 | histogram divisions per hour. The default is 1, and | |
266 | 0 suppresses histograms. Valid values are: | |
267 | ||
268 | 0, 1, 2, 3, 5, 10, 15, 20, 30 or 60. | |
269 | ||
270 | =item B<-ne> | |
271 | ||
272 | Don't display error information. | |
273 | ||
274 | =item B<-nr> | |
275 | ||
276 | Don't display relaying information. | |
277 | ||
278 | =item B<-nr>I</pattern/> | |
279 | ||
280 | Don't display relaying information that matches. | |
281 | ||
282 | =item B<-nt> | |
283 | ||
284 | Don't display transport information. | |
285 | ||
286 | =item B<-nt>I</pattern/> | |
287 | ||
288 | Don't display transport information that matches | |
289 | ||
290 | =item B<-q>I<list> | |
291 | ||
292 | List of times for queuing information single 0 item suppresses. | |
293 | ||
294 | =item B<-t>I<number> | |
295 | ||
296 | Display top <number> sources/destinations | |
297 | default is 50, 0 suppresses top listing. | |
298 | ||
299 | =item B<-tnl> | |
300 | ||
301 | Omit local sources/destinations in top listing. | |
302 | ||
303 | =item B<-t_remote_users> | |
304 | ||
305 | Include remote users in the top source/destination listings. | |
306 | ||
d5692f86 SC |
307 | =item B<-include_original_destination> |
308 | ||
309 | Include the original destination email addresses rather than just | |
310 | using the final ones. | |
311 | Useful for finding out which of your mailing lists are receiving mail. | |
312 | ||
059ec3d9 PH |
313 | =item B<-byhost> |
314 | ||
315 | Show results by sending host. This may be combined with | |
316 | B<-bydomain> and/or B<-byemail> and/or B<-byedomain>. If none of these options | |
317 | are specified, then B<-byhost> is assumed as a default. | |
318 | ||
319 | =item B<-bydomain> | |
320 | ||
321 | Show results by sending domain. | |
322 | May be combined with B<-byhost> and/or B<-byemail> and/or B<-byedomain>. | |
323 | ||
324 | =item B<-byemail> | |
325 | ||
326 | Show results by sender's email address. | |
327 | May be combined with B<-byhost> and/or B<-bydomain> and/or B<-byedomain>. | |
328 | ||
329 | =item B<-byemaildomain> or B<-byedomain> | |
330 | ||
331 | Show results by sender's email domain. | |
332 | May be combined with B<-byhost> and/or B<-bydomain> and/or B<-byemail>. | |
333 | ||
334 | =item B<-pattern> I<Description> I</Pattern/> | |
335 | ||
336 | Look for the specified pattern and count the number of lines in which it appears. | |
337 | This option can be specified multiple times. Eg: | |
338 | ||
339 | -pattern 'Refused connections' '/refused connection/' | |
340 | ||
341 | ||
342 | =item B<-merge> | |
343 | ||
344 | This option allows eximstats to merge old eximstat reports together. Eg: | |
345 | ||
346 | eximstats mainlog.sun > report.sun.txt | |
347 | eximstats mainlog.mon > report.mon.txt | |
348 | eximstats mainlog.tue > report.tue.txt | |
349 | eximstats mainlog.wed > report.web.txt | |
350 | eximstats mainlog.thu > report.thu.txt | |
351 | eximstats mainlog.fri > report.fri.txt | |
352 | eximstats mainlog.sat > report.sat.txt | |
353 | eximstats -merge report.*.txt > weekly_report.txt | |
354 | eximstats -merge -html report.*.txt > weekly_report.html | |
355 | ||
356 | =over 4 | |
357 | ||
358 | =item * | |
359 | ||
360 | You can merge text or html reports and output the results as text or html. | |
361 | ||
362 | =item * | |
363 | ||
364 | You can use all the normal eximstat output options, but only data | |
365 | included in the original reports can be shown! | |
366 | ||
367 | =item * | |
368 | ||
369 | When merging reports, some loss of accuracy may occur in the top I<n> lists. | |
370 | This will be towards the ends of the lists. | |
371 | ||
372 | =item * | |
373 | ||
374 | The order of items in the top I<n> lists may vary when the data volumes | |
375 | round to the same value. | |
376 | ||
377 | =back | |
378 | ||
059ec3d9 PH |
379 | =item B<-charts> |
380 | ||
381 | Create graphical charts to be displayed in HTML output. | |
d5692f86 | 382 | Only valid in combination with I<-html>. |
059ec3d9 PH |
383 | |
384 | This requires the following modules which can be obtained | |
385 | from http://www.cpan.org/modules/01modules.index.html | |
386 | ||
387 | =over 4 | |
388 | ||
389 | =item GD | |
390 | ||
391 | =item GDTextUtil | |
392 | ||
393 | =item GDGraph | |
394 | ||
395 | =back | |
396 | ||
397 | To install these, download and unpack them, then use the normal perl installation procedure: | |
398 | ||
399 | perl Makefile.PL | |
400 | make | |
401 | make test | |
402 | make install | |
403 | ||
404 | =item B<-chartdir>I <dir> | |
405 | ||
406 | Create the charts in the directory <dir> | |
407 | ||
408 | =item B<-chartrel>I <dir> | |
409 | ||
410 | Specify the relative directory for the "img src=" tags from where to include | |
411 | the charts | |
412 | ||
413 | =item B<-d> | |
414 | ||
415 | Debug flag. This outputs the eval()'d parser onto STDOUT which makes it | |
416 | easier to trap errors in the eval section. Remember to add 1 to the line numbers to allow for the | |
417 | title! | |
418 | ||
419 | =back | |
420 | ||
421 | =head1 DESCRIPTION | |
422 | ||
8974000d | 423 | Eximstats parses exim mainlog and syslog files to output a statistical |
059ec3d9 | 424 | analysis of the messages processed. By default, a text |
d5692f86 SC |
425 | analysis is generated, but you can request other output formats |
426 | using flags. See the help (B<-help>) to learn | |
059ec3d9 PH |
427 | about how to create charts from the tables. |
428 | ||
429 | =head1 AUTHOR | |
430 | ||
431 | There is a web site at http://www.exim.org - this contains details of the | |
432 | mailing list exim-users@exim.org. | |
433 | ||
434 | =head1 TO DO | |
435 | ||
436 | This program does not perfectly handle messages whose received | |
437 | and delivered log lines are in different files, which can happen | |
438 | when you have multiple mail servers and a message cannot be | |
439 | immeadiately delivered. Fixing this could be tricky... | |
440 | ||
d5692f86 SC |
441 | Merging of xls files is not (yet) possible. Be free to implement :) |
442 | ||
059ec3d9 PH |
443 | =cut |
444 | ||
445 | use integer; | |
446 | use strict; | |
d5692f86 | 447 | use IO::File; |
059ec3d9 PH |
448 | |
449 | # use Time::Local; # PH/FANF | |
450 | use POSIX; | |
451 | ||
d5692f86 | 452 | use vars qw($HAVE_GD_Graph_pie $HAVE_GD_Graph_linespoints $HAVE_Spreadsheet_WriteExcel); |
059ec3d9 PH |
453 | eval { require GD::Graph::pie; }; |
454 | $HAVE_GD_Graph_pie = $@ ? 0 : 1; | |
455 | eval { require GD::Graph::linespoints; }; | |
456 | $HAVE_GD_Graph_linespoints = $@ ? 0 : 1; | |
d5692f86 SC |
457 | eval { require Spreadsheet::WriteExcel; }; |
458 | $HAVE_Spreadsheet_WriteExcel = $@ ? 0 : 1; | |
059ec3d9 PH |
459 | |
460 | ||
461 | ################################################## | |
462 | # Static data # | |
463 | ################################################## | |
464 | # 'use vars' instead of 'our' as perl5.005 is still in use out there! | |
465 | use vars qw(@tab62 @days_per_month $gig); | |
466 | use vars qw($VERSION); | |
467 | use vars qw($COLUMN_WIDTHS); | |
468 | ||
469 | ||
470 | @tab62 = | |
471 | (0,1,2,3,4,5,6,7,8,9,0,0,0,0,0,0, # 0-9 | |
472 | 0,10,11,12,13,14,15,16,17,18,19,20, # A-K | |
473 | 21,22,23,24,25,26,27,28,29,30,31,32, # L-W | |
474 | 33,34,35, 0, 0, 0, 0, 0, # X-Z | |
475 | 0,36,37,38,39,40,41,42,43,44,45,46, # a-k | |
476 | 47,48,49,50,51,52,53,54,55,56,57,58, # l-w | |
477 | 59,60,61); # x-z | |
478 | ||
479 | @days_per_month = (0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334); | |
480 | $gig = 1024 * 1024 * 1024; | |
d5692f86 | 481 | $VERSION = '1.39'; |
059ec3d9 PH |
482 | |
483 | # How much space do we allow for the Hosts/Domains/Emails/Edomains column headers? | |
484 | $COLUMN_WIDTHS = 8; | |
485 | ||
486 | # Declare global variables. | |
487 | use vars qw($total_received_data $total_received_data_gigs $total_received_count); | |
488 | use vars qw($total_delivered_data $total_delivered_data_gigs $total_delivered_count); | |
489 | use vars qw(%arrival_time %size %from_host %from_address); | |
d5692f86 SC |
490 | use vars qw(%timestamp2time); #Hash of timestamp => time. |
491 | use vars qw($last_timestamp $last_time); #The last time convertion done. | |
492 | use vars qw($last_date $date_seconds); #The last date convertion done. | |
493 | use vars qw($last_offset $offset_seconds); #The last time offset convertion done. | |
059ec3d9 | 494 | use vars qw($localtime_offset); |
d5692f86 SC |
495 | use vars qw($i); #General loop counter. |
496 | use vars qw($debug); #Debug mode? | |
497 | use vars qw($ntopchart); #How many entries should make it into the chart? | |
498 | use vars qw($gddirectory); #Where to put files from GD::Graph | |
499 | use vars qw($workbook $ws_global $ws_relayed $ws_top50 $ws_errors ); #For use in Speadsheed::WriteExcel | |
500 | use vars qw($row $col $row_hist $col_hist $row_league_table); | |
501 | use vars qw($run_hist); | |
502 | use vars qw($f_default $f_header1 $f_header2 $f_headertab $f_percent); #Format Header | |
503 | ||
504 | # Output FileHandles | |
505 | use vars qw($txt_fh $htm_fh $xls_fh); | |
059ec3d9 PH |
506 | |
507 | $ntopchart = 5; | |
508 | ||
509 | # The following are parameters whose values are | |
510 | # set by command line switches: | |
511 | use vars qw($show_errors $show_relay $show_transport $transport_pattern); | |
512 | use vars qw($topcount $local_league_table $include_remote_users); | |
513 | use vars qw($hist_opt $hist_interval $hist_number $volume_rounding); | |
d5692f86 SC |
514 | use vars qw($relay_pattern @queue_times @user_patterns @user_descriptions); |
515 | use vars qw($include_original_destination); | |
516 | use vars qw($txt_fh $htm_fh $xls_fh); | |
059ec3d9 PH |
517 | |
518 | use vars qw(%do_sender); #Do sender by Host, Domain, Email, and/or Edomain tables. | |
519 | use vars qw($charts $chartrel $chartdir $charts_option_specified); | |
d5692f86 | 520 | use vars qw($merge_reports); #Merge old reports ? |
059ec3d9 PH |
521 | |
522 | # The following are modified in the parse() routine, and | |
523 | # referred to in the print_*() routines. | |
524 | use vars qw($queue_more_than $delayed_count $relayed_unshown $begin $end); | |
525 | use vars qw(%received_count %received_data %received_data_gigs); | |
526 | use vars qw(%delivered_count %delivered_data %delivered_data_gigs); | |
527 | use vars qw(%received_count_user %received_data_user %received_data_gigs_user); | |
528 | use vars qw(%delivered_count_user %delivered_data_user %delivered_data_gigs_user); | |
529 | use vars qw(%transported_count %transported_data %transported_data_gigs); | |
530 | use vars qw(%remote_delivered %relayed %delayed %had_error %errors_count); | |
531 | use vars qw(@queue_bin @remote_queue_bin @received_interval_count @delivered_interval_count); | |
532 | use vars qw(@user_pattern_totals); | |
533 | ||
534 | use vars qw(%report_totals); | |
535 | ||
536 | ||
537 | ||
538 | ||
539 | ################################################## | |
540 | # Subroutines # | |
541 | ################################################## | |
542 | ||
d5692f86 SC |
543 | ####################################################################### |
544 | # get_filehandle($file,\%output_files); | |
545 | # Return a filehandle writing to $file. | |
546 | # | |
547 | # If %output_files is defined, check that $output_files{$file} | |
548 | # doesn't exist and die if it does, or set it if it doesn't. | |
549 | ####################################################################### | |
550 | sub get_filehandle { | |
551 | my($file,$output_files_href) = @_; | |
552 | ||
553 | $file = '-' if ($file eq ''); | |
554 | ||
555 | if (defined $output_files_href) { | |
556 | die "You can only output to '$file' once! Use -h for help.\n" if exists $output_files_href->{$file}; | |
557 | $output_files_href->{$file} = 1; | |
558 | } | |
559 | ||
560 | if ($file eq '-') { | |
561 | return \*STDOUT; | |
562 | } | |
563 | ||
564 | if (-e $file) { | |
565 | unlink $file or die "Failed to rm $file: $!"; | |
566 | } | |
567 | ||
568 | my $fh = new IO::File $file, O_WRONLY|O_CREAT|O_EXCL; | |
569 | die "new IO::File $file failed: $!" unless (defined $fh); | |
570 | return $fh; | |
571 | } | |
572 | ||
059ec3d9 | 573 | |
1b4fe9dd PH |
574 | ####################################################################### |
575 | # volume_rounded(); | |
576 | # | |
577 | # $rounded_volume = volume_rounded($bytes,$gigabytes); | |
578 | # | |
579 | # Given a data size in bytes, round it to KB, MB, or GB | |
580 | # as appropriate. | |
581 | # | |
582 | # Eg 12000 => 12KB, 15000000 => 14GB, etc. | |
583 | # | |
584 | # Note: I've experimented with Math::BigInt and it results in a 33% | |
585 | # performance degredation as opposed to storing numbers split into | |
586 | # bytes and gigabytes. | |
587 | ####################################################################### | |
059ec3d9 PH |
588 | sub volume_rounded { |
589 | my($x,$g) = @_; | |
590 | $x = 0 unless $x; | |
591 | $g = 0 unless $g; | |
592 | my($rounded); | |
593 | ||
594 | while ($x > $gig) { | |
595 | $g++; | |
596 | $x -= $gig; | |
597 | } | |
598 | ||
599 | if ($volume_rounding) { | |
600 | # Values < 1 GB | |
601 | if ($g <= 0) { | |
602 | if ($x < 10000) { | |
603 | $rounded = sprintf("%6d", $x); | |
604 | } | |
605 | elsif ($x < 10000000) { | |
606 | $rounded = sprintf("%4dKB", ($x + 512)/1024); | |
607 | } | |
608 | else { | |
609 | $rounded = sprintf("%4dMB", ($x + 512*1024)/(1024*1024)); | |
610 | } | |
611 | } | |
612 | # Values between 1GB and 10GB are printed in MB | |
613 | elsif ($g < 10) { | |
614 | $rounded = sprintf("%4dMB", ($g * 1024) + ($x + 512*1024)/(1024*1024)); | |
615 | } | |
616 | else { | |
617 | # Handle values over 10GB | |
618 | $rounded = sprintf("%4dGB", $g + ($x + $gig/2)/$gig); | |
619 | } | |
620 | } | |
621 | else { | |
622 | # We don't want any rounding to be done. | |
d5692f86 SC |
623 | # and we don't need broken formated output which on one hand avoids numbers from |
624 | # being interpreted as string by Spreadsheed Calculators, on the other hand | |
625 | # breaks if more than 4 digits! -> flexible length instead of fixed length | |
626 | # Format the return value at the output routine! -fh | |
627 | #$rounded = sprintf("%d", ($g * $gig) + $x); | |
628 | no integer; | |
629 | $rounded = sprintf("%.0f", ($g * $gig) + $x); | |
059ec3d9 PH |
630 | } |
631 | ||
632 | return $rounded; | |
633 | } | |
634 | ||
635 | ||
1b4fe9dd PH |
636 | ####################################################################### |
637 | # un_round(); | |
8e669ac1 | 638 | # |
1b4fe9dd | 639 | # un_round($rounded_volume,\$bytes,\$gigabytes); |
8e669ac1 | 640 | # |
1b4fe9dd PH |
641 | # Given a volume in KB, MB or GB, as generated by volume_rounded(), |
642 | # do the reverse transformation and convert it back into Bytes and Gigabytes. | |
643 | # These are added to the $bytes and $gigabytes parameters. | |
8e669ac1 | 644 | # |
1b4fe9dd PH |
645 | # Given a data size in bytes, round it to KB, MB, or GB |
646 | # as appropriate. | |
8e669ac1 | 647 | # |
1b4fe9dd PH |
648 | # EG: 500 => (500,0), 14GB => (0,14), etc. |
649 | ####################################################################### | |
059ec3d9 PH |
650 | sub un_round { |
651 | my($rounded,$bytes_sref,$gigabytes_sref) = @_; | |
652 | ||
653 | if ($rounded =~ /(\d+)GB/) { | |
654 | $$gigabytes_sref += $1; | |
655 | } | |
656 | elsif ($rounded =~ /(\d+)MB/) { | |
657 | $$gigabytes_sref += $1 / 1024; | |
658 | $$bytes_sref += (($1 % 1024 ) * 1024 * 1024); | |
659 | } | |
660 | elsif ($rounded =~ /(\d+)KB/) { | |
661 | $$gigabytes_sref += $1 / (1024 * 1024); | |
662 | $$bytes_sref += ($1 % (1024 * 1024) * 1024); | |
663 | } | |
664 | elsif ($rounded =~ /(\d+)/) { | |
d5692f86 SC |
665 | # We need to turn off integer in case we are merging an -nvr report. |
666 | no integer; | |
667 | $$gigabytes_sref += int($1 / $gig); | |
059ec3d9 PH |
668 | $$bytes_sref += $1 % $gig; |
669 | } | |
670 | ||
671 | #Now reduce the bytes down to less than 1GB. | |
672 | add_volume($bytes_sref,$gigabytes_sref,0) if ($$bytes_sref > $gig); | |
673 | } | |
674 | ||
675 | ||
1b4fe9dd PH |
676 | ####################################################################### |
677 | # add_volume(); | |
8e669ac1 | 678 | # |
1b4fe9dd | 679 | # add_volume(\$bytes,\$gigs,$size); |
8e669ac1 | 680 | # |
1b4fe9dd PH |
681 | # Add $size to $bytes/$gigs where this is a number split into |
682 | # bytes ($bytes) and gigabytes ($gigs). This is significantly | |
683 | # faster than using Math::BigInt. | |
684 | ####################################################################### | |
059ec3d9 | 685 | sub add_volume { |
1b4fe9dd PH |
686 | my($bytes_ref,$gigs_ref,$size) = @_; |
687 | $$bytes_ref = 0 if ! defined $$bytes_ref; | |
688 | $$gigs_ref = 0 if ! defined $$gigs_ref; | |
689 | $$bytes_ref += $size; | |
690 | while ($$bytes_ref > $gig) { | |
691 | $$gigs_ref++; | |
692 | $$bytes_ref -= $gig; | |
059ec3d9 PH |
693 | } |
694 | } | |
695 | ||
696 | ||
1b4fe9dd PH |
697 | ####################################################################### |
698 | # format_time(); | |
8e669ac1 | 699 | # |
1b4fe9dd | 700 | # $formatted_time = format_time($seconds); |
8e669ac1 | 701 | # |
1b4fe9dd PH |
702 | # Given a time in seconds, break it down into |
703 | # weeks, days, hours, minutes, and seconds. | |
8e669ac1 | 704 | # |
1b4fe9dd PH |
705 | # Eg 12005 => 3h20m5s |
706 | ####################################################################### | |
059ec3d9 PH |
707 | sub format_time { |
708 | my($t) = pop @_; | |
709 | my($s) = $t % 60; | |
710 | $t /= 60; | |
711 | my($m) = $t % 60; | |
712 | $t /= 60; | |
713 | my($h) = $t % 24; | |
714 | $t /= 24; | |
715 | my($d) = $t % 7; | |
716 | my($w) = $t/7; | |
717 | my($p) = ""; | |
718 | $p .= "$w"."w" if $w > 0; | |
719 | $p .= "$d"."d" if $d > 0; | |
720 | $p .= "$h"."h" if $h > 0; | |
721 | $p .= "$m"."m" if $m > 0; | |
722 | $p .= "$s"."s" if $s > 0 || $p eq ""; | |
723 | $p; | |
724 | } | |
725 | ||
726 | ||
1b4fe9dd PH |
727 | ####################################################################### |
728 | # unformat_time(); | |
8e669ac1 | 729 | # |
1b4fe9dd | 730 | # $seconds = unformat_time($formatted_time); |
8e669ac1 | 731 | # |
1b4fe9dd | 732 | # Given a time in weeks, days, hours, minutes, or seconds, convert it to seconds. |
8e669ac1 | 733 | # |
1b4fe9dd PH |
734 | # Eg 3h20m5s => 12005 |
735 | ####################################################################### | |
059ec3d9 PH |
736 | sub unformat_time { |
737 | my($formated_time) = pop @_; | |
738 | my $time = 0; | |
739 | ||
740 | while ($formated_time =~ s/^(\d+)([wdhms]?)//) { | |
741 | $time += $1 if ($2 eq '' || $2 eq 's'); | |
742 | $time += $1 * 60 if ($2 eq 'm'); | |
743 | $time += $1 * 60 * 60 if ($2 eq 'h'); | |
744 | $time += $1 * 60 * 60 * 24 if ($2 eq 'd'); | |
745 | $time += $1 * 60 * 60 * 24 * 7 if ($2 eq 'w'); | |
746 | } | |
747 | $time; | |
748 | } | |
749 | ||
750 | ||
1b4fe9dd PH |
751 | ####################################################################### |
752 | # seconds(); | |
8e669ac1 | 753 | # |
1b4fe9dd | 754 | # $time = seconds($timestamp); |
8e669ac1 | 755 | # |
1b4fe9dd PH |
756 | # Given a time-of-day timestamp, convert it into a time() value using |
757 | # POSIX::mktime. We expect the timestamp to be of the form | |
758 | # "$year-$mon-$day $hour:$min:$sec", with month going from 1 to 12, | |
759 | # and the year to be absolute (we do the necessary conversions). The | |
760 | # timestamp may be followed with an offset from UTC like "+$hh$mm"; if the | |
761 | # offset is not present, and we have not been told that the log is in UTC | |
762 | # (with the -utc option), then we adjust the time by the current local | |
763 | # time offset so that it can be compared with the time recorded in message | |
764 | # IDs, which is UTC. | |
8e669ac1 | 765 | # |
1b4fe9dd PH |
766 | # To improve performance, we only use mktime on the date ($year-$mon-$day), |
767 | # and only calculate it if the date is different to the previous time we | |
768 | # came here. We then add on seconds for the '$hour:$min:$sec'. | |
8e669ac1 | 769 | # |
1b4fe9dd PH |
770 | # We also store the results of the last conversion done, and only |
771 | # recalculate if the date is different. | |
8e669ac1 | 772 | # |
1b4fe9dd PH |
773 | # We used to have the '-cache' flag which would store the results of the |
774 | # mktime() call. However, the current way of just using mktime() on the | |
775 | # date obsoletes this. | |
776 | ####################################################################### | |
059ec3d9 PH |
777 | sub seconds { |
778 | my($timestamp) = @_; | |
779 | ||
780 | # Is the timestamp the same as the last one? | |
781 | return $last_time if ($last_timestamp eq $timestamp); | |
782 | ||
783 | return 0 unless ($timestamp =~ /^((\d{4})\-(\d\d)-(\d\d))\s(\d\d):(\d\d):(\d\d)( ([+-])(\d\d)(\d\d))?/o); | |
784 | ||
785 | unless ($last_date eq $1) { | |
786 | $last_date = $1; | |
787 | my(@timestamp) = (0,0,0,$4,$3,$2); | |
788 | $timestamp[5] -= 1900; | |
789 | $timestamp[4]--; | |
790 | $date_seconds = mktime(@timestamp); | |
791 | } | |
792 | my $time = $date_seconds + ($5 * 3600) + ($6 * 60) + $7; | |
793 | ||
794 | # SC. Use cacheing. Also note we want seconds not minutes. | |
795 | #my($this_offset) = ($10 * 60 + $11) * ($9 . "1") if defined $8; | |
796 | if (defined $8 && ($8 ne $last_offset)) { | |
797 | $last_offset = $8; | |
798 | $offset_seconds = ($10 * 60 + $11) * 60; | |
799 | $offset_seconds = -$offset_seconds if ($9 eq '-'); | |
800 | } | |
801 | ||
802 | ||
803 | if (defined $7) { | |
804 | #$time -= $this_offset; | |
805 | $time -= $offset_seconds; | |
806 | } elsif (defined $localtime_offset) { | |
807 | $time -= $localtime_offset; | |
808 | } | |
809 | ||
810 | # Store the last timestamp received. | |
811 | $last_timestamp = $timestamp; | |
812 | $last_time = $time; | |
813 | ||
814 | $time; | |
815 | } | |
816 | ||
817 | ||
1b4fe9dd PH |
818 | ####################################################################### |
819 | # id_seconds(); | |
8e669ac1 | 820 | # |
1b4fe9dd | 821 | # $time = id_seconds($message_id); |
8e669ac1 | 822 | # |
1b4fe9dd PH |
823 | # Given a message ID, convert it into a time() value. |
824 | ####################################################################### | |
059ec3d9 PH |
825 | sub id_seconds { |
826 | my($sub_id) = substr((pop @_), 0, 6); | |
827 | my($s) = 0; | |
828 | my(@c) = split(//, $sub_id); | |
829 | while($#c >= 0) { $s = $s * 62 + $tab62[ord(shift @c) - ord('0')] } | |
830 | $s; | |
831 | } | |
832 | ||
833 | ||
834 | ||
1b4fe9dd PH |
835 | ####################################################################### |
836 | # calculate_localtime_offset(); | |
8e669ac1 | 837 | # |
1b4fe9dd | 838 | # $localtime_offset = calculate_localtime_offset(); |
8e669ac1 | 839 | # |
1b4fe9dd | 840 | # Calculate the the localtime offset from gmtime in seconds. |
8e669ac1 | 841 | # |
1b4fe9dd | 842 | # $localtime = time() + $localtime_offset. |
8e669ac1 | 843 | # |
1b4fe9dd PH |
844 | # These are the same semantics as ISO 8601 and RFC 2822 timezone offsets. |
845 | # (West is negative, East is positive.) | |
846 | ####################################################################### | |
059ec3d9 PH |
847 | |
848 | # $localtime = gmtime() + $localtime_offset. OLD COMMENT | |
849 | # This subroutine commented out as it's not currently in use. | |
850 | ||
851 | #sub calculate_localtime_offset { | |
852 | # # Pick an arbitrary date, convert it to localtime & gmtime, and return the difference. | |
853 | # my (@sample_date) = (0,0,0,5,5,100); | |
854 | # my $localtime = timelocal(@sample_date); | |
855 | # my $gmtime = timegm(@sample_date); | |
856 | # my $offset = $localtime - $gmtime; | |
857 | # return $offset; | |
858 | #} | |
859 | ||
860 | sub calculate_localtime_offset { | |
861 | # Assume that the offset at the moment is valid across the whole | |
862 | # period covered by the logs that we're analysing. This may not | |
863 | # be true around the time the clocks change in spring or autumn. | |
864 | my $utc = time; | |
865 | # mktime works on local time and gmtime works in UTC | |
866 | my $local = mktime(gmtime($utc)); | |
867 | return $local - $utc; | |
868 | } | |
869 | ||
870 | ||
1b4fe9dd PH |
871 | ####################################################################### |
872 | # print_queue_times(); | |
8e669ac1 | 873 | # |
1b4fe9dd | 874 | # $time = print_queue_times($message_type,\@queue_times,$queue_more_than); |
8e669ac1 | 875 | # |
1b4fe9dd PH |
876 | # Given the type of messages being output, the array of message queue times, |
877 | # and the number of messages which exceeded the queue times, print out | |
878 | # a table. | |
879 | ####################################################################### | |
059ec3d9 PH |
880 | sub print_queue_times { |
881 | no integer; | |
882 | my($string,$array,$queue_more_than) = @_; | |
883 | my(@chartdatanames); | |
884 | my(@chartdatavals); | |
885 | ||
886 | my $printed_one = 0; | |
887 | my $cumulative_percent = 0; | |
888 | #$queue_unknown += keys %arrival_time; | |
889 | ||
890 | my $queue_total = $queue_more_than; | |
891 | for ($i = 0; $i <= $#queue_times; $i++) { $queue_total += $$array[$i] } | |
892 | ||
893 | my $temp = "Time spent on the queue: $string"; | |
894 | ||
d5692f86 SC |
895 | |
896 | my $txt_format = "%5s %4s %6d %5.1f%% %5.1f%%\n"; | |
897 | my $htm_format = "<tr><td align=\"right\">%s %s</td><td align=\"right\">%d</td><td align=\"right\">%5.1f%%</td><td align=\"right\">%5.1f%%</td>\n"; | |
898 | ||
899 | # write header | |
900 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; | |
901 | if ($htm_fh) { | |
902 | print $htm_fh "<hr><a name=\"$string time\"></a><h2>$temp</h2>\n"; | |
903 | print $htm_fh "<table border=0 width=\"100%\">\n"; | |
904 | print $htm_fh "<tr><td>\n"; | |
905 | print $htm_fh "<table border=1>\n"; | |
906 | print $htm_fh "<tr><th>Time</th><th>Messages</th><th>Percentage</th><th>Cumulative Percentage</th>\n"; | |
059ec3d9 | 907 | } |
d5692f86 | 908 | if ($xls_fh) |
059ec3d9 | 909 | { |
d5692f86 SC |
910 | |
911 | $ws_global->write($row++, $col, "Time spent on the queue: ".$string, $f_header2); | |
912 | my @content=("Time", "Messages", "Percentage", "Cumulative Percentage"); | |
913 | &set_worksheet_line($ws_global, $row++, 1, \@content, $f_headertab); | |
059ec3d9 PH |
914 | } |
915 | ||
d5692f86 | 916 | |
059ec3d9 PH |
917 | for ($i = 0; $i <= $#queue_times; $i++) { |
918 | if ($$array[$i] > 0) | |
919 | { | |
920 | my $percent = ($$array[$i] * 100)/$queue_total; | |
921 | $cumulative_percent += $percent; | |
d5692f86 SC |
922 | |
923 | my @content=($printed_one? " " : "Under", | |
924 | format_time($queue_times[$i]), | |
925 | $$array[$i], $percent, $cumulative_percent); | |
926 | ||
927 | if ($htm_fh) { | |
928 | printf $htm_fh ($htm_format, @content); | |
929 | if (!defined($queue_times[$i])) { | |
930 | print $htm_fh "Not defined"; | |
931 | } | |
932 | } | |
933 | if ($txt_fh) { | |
934 | printf $txt_fh ($txt_format, @content); | |
935 | if (!defined($queue_times[$i])) { | |
936 | print $txt_fh "Not defined"; | |
937 | } | |
938 | } | |
939 | if ($xls_fh) | |
940 | { | |
941 | no integer; | |
942 | &set_worksheet_line($ws_global, $row, 0, [@content[0,1,2]], $f_default); | |
943 | &set_worksheet_line($ws_global, $row++, 3, [$content[3]/100,$content[4]/100], $f_percent); | |
944 | ||
945 | if (!defined($queue_times[$i])) { | |
946 | $col=0; | |
947 | $ws_global->write($row++, $col, "Not defined" ); | |
948 | } | |
059ec3d9 | 949 | } |
d5692f86 | 950 | |
059ec3d9 PH |
951 | push(@chartdatanames, |
952 | ($printed_one? "" : "Under") . format_time($queue_times[$i])); | |
953 | push(@chartdatavals, $$array[$i]); | |
954 | $printed_one = 1; | |
955 | } | |
956 | } | |
957 | ||
958 | if ($queue_more_than > 0) { | |
959 | my $percent = ($queue_more_than * 100)/$queue_total; | |
960 | $cumulative_percent += $percent; | |
d5692f86 SC |
961 | |
962 | my @content = ("Over ", format_time($queue_times[$#queue_times]), | |
963 | $queue_more_than, $percent, $cumulative_percent); | |
964 | ||
965 | printf $txt_fh ($txt_format, @content) if $txt_fh; | |
966 | printf $htm_fh ($htm_format, @content) if $htm_fh; | |
967 | if ($xls_fh) | |
968 | { | |
969 | &set_worksheet_line($ws_global, $row, 0, [@content[0,1,2]], $f_default); | |
970 | &set_worksheet_line($ws_global, $row++, 3, [$content[3]/100,$content[4]/100], $f_percent); | |
971 | } | |
972 | ||
059ec3d9 | 973 | } |
d5692f86 | 974 | |
059ec3d9 PH |
975 | push(@chartdatanames, "Over " . format_time($queue_times[$#queue_times])); |
976 | push(@chartdatavals, $queue_more_than); | |
977 | ||
978 | #printf("Unknown %6d\n", $queue_unknown) if $queue_unknown > 0; | |
d5692f86 SC |
979 | if ($htm_fh) { |
980 | print $htm_fh "</table>\n"; | |
981 | print $htm_fh "</td><td>\n"; | |
059ec3d9 PH |
982 | |
983 | if ($HAVE_GD_Graph_pie && $charts) { | |
984 | my @data = ( | |
985 | \@chartdatanames, | |
986 | \@chartdatavals | |
987 | ); | |
988 | my $graph = GD::Graph::pie->new(200, 200); | |
989 | my $pngname; | |
990 | my $title; | |
991 | if ($string =~ /all/) { $pngname = "queue_all.png"; $title = "Queue (all)"; } | |
992 | if ($string =~ /remote/) { $pngname = "queue_rem.png"; $title = "Queue (remote)"; } | |
993 | $graph->set( | |
994 | title => $title, | |
995 | ); | |
996 | my $gd = $graph->plot(\@data) or warn($graph->error); | |
997 | if ($gd) { | |
d5692f86 | 998 | open(IMG, ">$chartdir/$pngname") or die "Could not write $chartdir/$pngname: $!\n"; |
059ec3d9 PH |
999 | binmode IMG; |
1000 | print IMG $gd->png; | |
1001 | close IMG; | |
d5692f86 | 1002 | print $htm_fh "<img src=\"$chartrel/$pngname\">"; |
059ec3d9 PH |
1003 | } |
1004 | } | |
d5692f86 SC |
1005 | print $htm_fh "</td></tr></table>\n"; |
1006 | } | |
1007 | ||
1008 | if ($xls_fh) | |
1009 | { | |
1010 | $row++; | |
059ec3d9 | 1011 | } |
d5692f86 SC |
1012 | print $txt_fh "\n" if $txt_fh; |
1013 | print $htm_fh "\n" if $htm_fh; | |
1014 | ||
059ec3d9 PH |
1015 | } |
1016 | ||
1017 | ||
1018 | ||
1b4fe9dd PH |
1019 | ####################################################################### |
1020 | # print_histogram(); | |
8e669ac1 | 1021 | # |
1b4fe9dd | 1022 | # print_histogram('Deliverieds|Messages received',@interval_count); |
8e669ac1 | 1023 | # |
1b4fe9dd PH |
1024 | # Print a histogram of the messages delivered/received per time slot |
1025 | # (hour by default). | |
1026 | ####################################################################### | |
059ec3d9 PH |
1027 | sub print_histogram { |
1028 | my($text) = shift; | |
1029 | my(@interval_count) = @_; | |
1030 | my(@chartdatanames); | |
1031 | my(@chartdatavals); | |
1032 | my($maxd) = 0; | |
d5692f86 SC |
1033 | |
1034 | if (!$run_hist) # save first row of print_histogram for xls output | |
1035 | { | |
1036 | $row_hist = $row; | |
1037 | } | |
1038 | else | |
1039 | { | |
1040 | $row = $row_hist; | |
1041 | } | |
1042 | ||
059ec3d9 PH |
1043 | for ($i = 0; $i < $hist_number; $i++) |
1044 | { $maxd = $interval_count[$i] if $interval_count[$i] > $maxd; } | |
1045 | ||
1046 | my $scale = int(($maxd + 25)/50); | |
1047 | $scale = 1 if $scale == 0; | |
1048 | ||
1049 | my($type); | |
1050 | if ($text eq "Deliveries") | |
d5692f86 | 1051 | { |
059ec3d9 | 1052 | $type = ($scale == 1)? "delivery" : "deliveries"; |
d5692f86 | 1053 | } |
059ec3d9 | 1054 | else |
d5692f86 | 1055 | { |
059ec3d9 | 1056 | $type = ($scale == 1)? "message" : "messages"; |
d5692f86 SC |
1057 | } |
1058 | ||
1059 | # make and output title | |
1060 | my $title = sprintf("$text per %s", | |
1061 | ($hist_interval == 60)? "hour" : | |
1062 | ($hist_interval == 1)? "minute" : "$hist_interval minutes"); | |
059ec3d9 | 1063 | |
d5692f86 | 1064 | my $txt_htm_title = $title . " (each dot is $scale $type)"; |
059ec3d9 | 1065 | |
d5692f86 SC |
1066 | printf $txt_fh ("%s\n%s\n\n", $txt_htm_title, "-" x length($txt_htm_title)) if $txt_fh; |
1067 | ||
1068 | if ($htm_fh) { | |
1069 | print $htm_fh "<hr><a name=\"$text\"></a><h2>$txt_htm_title</h2>\n"; | |
1070 | print $htm_fh "<table border=0 width=\"100%\">\n"; | |
1071 | print $htm_fh "<tr><td><pre>\n"; | |
059ec3d9 | 1072 | } |
d5692f86 SC |
1073 | |
1074 | if ($xls_fh) | |
1075 | { | |
1076 | $title =~ s/Messages/Msg/ ; | |
1077 | $ws_global->write($row++, $col_hist+1, $title, $f_headertab); | |
059ec3d9 PH |
1078 | } |
1079 | ||
d5692f86 | 1080 | |
059ec3d9 PH |
1081 | my $hour = 0; |
1082 | my $minutes = 0; | |
1083 | for ($i = 0; $i < $hist_number; $i++) | |
d5692f86 | 1084 | { |
059ec3d9 PH |
1085 | my $c = $interval_count[$i]; |
1086 | ||
1087 | # If the interval is an hour (the maximum) print the starting and | |
1088 | # ending hours as a label. Otherwise print the starting hour and | |
1089 | # minutes, which take up the same space. | |
1090 | ||
1091 | my $temp; | |
1092 | if ($hist_opt == 1) | |
d5692f86 | 1093 | { |
059ec3d9 | 1094 | $temp = sprintf("%02d-%02d", $hour, $hour + 1); |
d5692f86 SC |
1095 | |
1096 | print $txt_fh $temp if $txt_fh; | |
1097 | print $htm_fh $temp if $htm_fh; | |
1098 | ||
1099 | if ($xls_fh) | |
1100 | { | |
1101 | if ($run_hist==0) # only on first run | |
1102 | { | |
1103 | &set_worksheet_line($ws_global, $row, 0, [$temp], $f_default); | |
1104 | } | |
1105 | } | |
1106 | ||
059ec3d9 PH |
1107 | push(@chartdatanames, $temp); |
1108 | $hour++; | |
d5692f86 | 1109 | } |
059ec3d9 | 1110 | else |
d5692f86 | 1111 | { |
059ec3d9 PH |
1112 | if ($minutes == 0) |
1113 | { $temp = sprintf("%02d:%02d", $hour, $minutes) } | |
1114 | else | |
1115 | { $temp = sprintf(" :%02d", $minutes) } | |
d5692f86 SC |
1116 | |
1117 | print $txt_fh $temp if $txt_fh; | |
1118 | print $htm_fh $temp if $htm_fh; | |
1119 | if (($xls_fh) and ($run_hist==0)) # only on first run | |
1120 | { | |
1121 | $temp = sprintf("%02d:%02d", $hour, $minutes); | |
1122 | &set_worksheet_line($ws_global, $row, 0, [$temp], $f_default); | |
1123 | } | |
1124 | ||
059ec3d9 PH |
1125 | push(@chartdatanames, $temp); |
1126 | $minutes += $hist_interval; | |
1127 | if ($minutes >= 60) | |
1128 | { | |
1129 | $minutes = 0; | |
1130 | $hour++; | |
1131 | } | |
059ec3d9 | 1132 | } |
d5692f86 SC |
1133 | push(@chartdatavals, $c); |
1134 | ||
1135 | printf $txt_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $txt_fh; | |
1136 | printf $htm_fh (" %6d %s\n", $c, "." x ($c/$scale)) if $htm_fh; | |
1137 | if ($xls_fh) | |
059ec3d9 | 1138 | { |
d5692f86 SC |
1139 | &set_worksheet_line($ws_global, $row++, $col_hist+1, [$c], $f_default); |
1140 | } | |
1141 | ||
1142 | } #end for | |
1143 | ||
1144 | printf $txt_fh "\n" if $txt_fh; | |
1145 | printf $htm_fh "\n" if $htm_fh; | |
1146 | ||
1147 | if ($htm_fh) | |
1148 | { | |
1149 | print $htm_fh "</pre>\n"; | |
1150 | print $htm_fh "</td><td>\n"; | |
059ec3d9 PH |
1151 | if ($HAVE_GD_Graph_linespoints && $charts) { |
1152 | # calculate the graph | |
1153 | my @data = ( | |
1154 | \@chartdatanames, | |
1155 | \@chartdatavals | |
1156 | ); | |
1157 | my $graph = GD::Graph::linespoints->new(300, 300); | |
1158 | $graph->set( | |
1159 | x_label => 'Time', | |
1160 | y_label => 'Amount', | |
1161 | title => $text, | |
1162 | x_labels_vertical => 1 | |
1163 | ); | |
1164 | my($pngname); | |
1165 | if ($text =~ /Deliveries/) { $pngname = "histogram_del.png"; } | |
1166 | if ($text =~ /Messages/) { $pngname = "histogram_mes.png"; } | |
1167 | my $gd = $graph->plot(\@data) or warn($graph->error); | |
1168 | if ($gd) { | |
d5692f86 | 1169 | open(IMG, ">$chartdir/$pngname") or die "Could not write $chartdir/$pngname: $!\n"; |
059ec3d9 PH |
1170 | binmode IMG; |
1171 | print IMG $gd->png; | |
1172 | close IMG; | |
d5692f86 | 1173 | print $htm_fh "<img src=\"$chartrel/$pngname\">"; |
059ec3d9 PH |
1174 | } |
1175 | } | |
d5692f86 | 1176 | print $htm_fh "</td></tr></table>\n"; |
059ec3d9 | 1177 | } |
d5692f86 SC |
1178 | |
1179 | $col_hist++; # where to continue next times | |
1180 | ||
1181 | $row+=2; # leave some space after history block | |
1182 | $run_hist=1; # we have done this once or more | |
059ec3d9 PH |
1183 | } |
1184 | ||
1185 | ||
1186 | ||
1b4fe9dd PH |
1187 | ####################################################################### |
1188 | # print_league_table(); | |
8e669ac1 | 1189 | # |
1b4fe9dd | 1190 | # print_league_table($league_table_type,\%message_count,\%message_data,\%message_data_gigs); |
8e669ac1 | 1191 | # |
1b4fe9dd PH |
1192 | # Given hashes of message count and message data, which are keyed by |
1193 | # the table type (eg by the sending host), print a league table | |
1194 | # showing the top $topcount (defaults to 50). | |
1195 | ####################################################################### | |
059ec3d9 PH |
1196 | sub print_league_table { |
1197 | my($text,$m_count,$m_data,$m_data_gigs) = @_; | |
1198 | my($name) = ($topcount == 1)? "$text" : "$topcount ${text}s"; | |
1199 | my($temp) = "Top $name by message count"; | |
1200 | my(@chartdatanames) = (); | |
1201 | my(@chartdatavals) = (); | |
1202 | my $chartotherval = 0; | |
1203 | ||
d5692f86 SC |
1204 | my $htm_format; |
1205 | my $txt_format = "%7d %10s %s\n"; | |
1206 | ||
1207 | # write header | |
1208 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; | |
1209 | if ($htm_fh) { | |
1210 | print $htm_fh "<hr><a name=\"$text count\"></a><h2>$temp</h2>\n"; | |
1211 | print $htm_fh "<table border=0 width=\"100%\">\n"; | |
1212 | print $htm_fh "<tr><td>\n"; | |
1213 | print $htm_fh "<table border=1>\n"; | |
1214 | print $htm_fh "<tr><th>Messages</th><th>Bytes</th><th>Average</th><th>\u$text</th>\n"; | |
059ec3d9 PH |
1215 | |
1216 | # Align non-local addresses to the right (so all the .com's line up). | |
1217 | # Local addresses are aligned on the left as they are userids. | |
1218 | my $align = ($text !~ /local/i) ? 'right' : 'left'; | |
d5692f86 | 1219 | $htm_format = "<tr><td align=\"right\">%d</td><td align=\"right\">%s</td><td align=\"right\">%s</td><td align=\"$align\" nowrap>%s</td>\n"; |
059ec3d9 | 1220 | } |
d5692f86 SC |
1221 | if ($xls_fh) |
1222 | { | |
1223 | $ws_top50->write($row_league_table++, 0, $temp, $f_header2); | |
1224 | &set_worksheet_line($ws_top50, $row_league_table++, 0, ["Messages", "Bytes", "Average", $text], $f_headertab ); | |
059ec3d9 PH |
1225 | } |
1226 | ||
d5692f86 SC |
1227 | |
1228 | # write content | |
50adf73a | 1229 | my($key,$htmlkey,$rounded_volume,$rounded_average,$count,$data,$gigs); |
059ec3d9 | 1230 | foreach $key (top_n_sort($topcount,$m_count,$m_data_gigs,$m_data)) { |
d5692f86 SC |
1231 | |
1232 | # When displaying the average figures, we calculate the average of | |
1233 | # the rounded data, as the user would calculate it. This reduces | |
1234 | # the accuracy slightly, but we have to do it this way otherwise | |
1235 | # when using -merge to convert results from text to HTML and | |
1236 | # vice-versa discrepencies would occur. | |
1237 | $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); | |
1238 | $data = $gigs = 0; | |
1239 | un_round($rounded_volume,\$data,\$gigs); | |
1240 | $count = $$m_count{$key}; | |
1241 | $rounded_average = volume_rounded($data/$count,$gigs/$count); | |
1242 | my @content=( $count, $rounded_volume, $rounded_average); | |
1243 | ||
1244 | # write content | |
1245 | # any reason not to include rounded_average in txt-output? -fh | |
1246 | printf $txt_fh ($txt_format, $count, $rounded_volume, $key) if $txt_fh; | |
1247 | ||
1248 | if ($htm_fh) { | |
059ec3d9 PH |
1249 | $htmlkey = $key; |
1250 | $htmlkey =~ s/>/\>\;/g; | |
1251 | $htmlkey =~ s/</\<\;/g; | |
d5692f86 | 1252 | printf $htm_fh ($htm_format, @content, $htmlkey); |
059ec3d9 | 1253 | } |
d5692f86 SC |
1254 | if ($xls_fh) |
1255 | { | |
1256 | &set_worksheet_line($ws_top50, $row_league_table++, 0, [@content, $key], $f_default); | |
059ec3d9 | 1257 | } |
d5692f86 | 1258 | |
059ec3d9 | 1259 | if (scalar @chartdatanames < $ntopchart) |
d5692f86 | 1260 | { |
059ec3d9 PH |
1261 | push(@chartdatanames, $key); |
1262 | push(@chartdatavals, $$m_count{$key}); | |
d5692f86 | 1263 | } |
059ec3d9 | 1264 | else |
d5692f86 | 1265 | { |
059ec3d9 | 1266 | $chartotherval += $$m_count{$key}; |
059ec3d9 | 1267 | } |
d5692f86 SC |
1268 | } |
1269 | ||
059ec3d9 PH |
1270 | push(@chartdatanames, "Other"); |
1271 | push(@chartdatavals, $chartotherval); | |
1272 | ||
d5692f86 SC |
1273 | print $txt_fh "\n" if $txt_fh; |
1274 | if ($htm_fh) | |
1275 | { | |
1276 | print $htm_fh "</table>\n"; | |
1277 | print $htm_fh "</td><td>\n"; | |
059ec3d9 PH |
1278 | if ($HAVE_GD_Graph_pie && $charts) |
1279 | { | |
1280 | # calculate the graph | |
1281 | my @data = ( | |
1282 | \@chartdatanames, | |
1283 | \@chartdatavals | |
1284 | ); | |
1285 | my $graph = GD::Graph::pie->new(300, 300); | |
1286 | $graph->set( | |
1287 | x_label => 'Name', | |
1288 | y_label => 'Amount', | |
1289 | title => 'By count', | |
1290 | ); | |
1291 | my $gd = $graph->plot(\@data) or warn($graph->error); | |
1292 | if ($gd) { | |
1293 | my $temp = $text; | |
1294 | $temp =~ s/ /_/g; | |
d5692f86 | 1295 | open(IMG, ">$chartdir/${temp}_count.png") or die "Could not write $chartdir/${temp}_count.png: $!\n"; |
059ec3d9 PH |
1296 | binmode IMG; |
1297 | print IMG $gd->png; | |
1298 | close IMG; | |
d5692f86 | 1299 | print $htm_fh "<img src=\"$chartrel/${temp}_count.png\">"; |
059ec3d9 PH |
1300 | } |
1301 | } | |
d5692f86 SC |
1302 | print $htm_fh "</td><td>\n"; |
1303 | print $htm_fh "</td></tr></table>\n\n"; | |
1304 | } | |
1305 | if ($xls_fh) | |
1306 | { | |
1307 | $row_league_table++; | |
059ec3d9 | 1308 | } |
d5692f86 SC |
1309 | |
1310 | ||
1311 | # write header | |
059ec3d9 PH |
1312 | |
1313 | $temp = "Top $name by volume"; | |
d5692f86 SC |
1314 | |
1315 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; | |
1316 | if ($htm_fh) { | |
1317 | print $htm_fh "<hr><a name=\"$text volume\"></a><h2>$temp</h2>\n"; | |
1318 | print $htm_fh "<table border=0 width=\"100%\">\n"; | |
1319 | print $htm_fh "<tr><td>\n"; | |
1320 | print $htm_fh "<table border=1>\n"; | |
1321 | print $htm_fh "<tr><th>Messages</th><th>Bytes</th><th>Average</th><th>\u$text</th>\n"; | |
059ec3d9 | 1322 | } |
d5692f86 SC |
1323 | if ($xls_fh) |
1324 | { | |
1325 | $ws_top50->write($row_league_table++, 0, $temp, $f_header2); | |
1326 | &set_worksheet_line($ws_top50, $row_league_table++, 0, ["Messages", "Bytes", "Average", $text], $f_headertab); | |
059ec3d9 PH |
1327 | } |
1328 | ||
1329 | @chartdatanames = (); | |
1330 | @chartdatavals = (); | |
1331 | $chartotherval = 0; | |
8974000d | 1332 | my $use_gig = 0; |
059ec3d9 | 1333 | foreach $key (top_n_sort($topcount,$m_data_gigs,$m_data,$m_count)) { |
884c2af8 SC |
1334 | # The largest volume will be the first (top of the list). |
1335 | # If it has at least 1 gig, then just use gigabytes to avoid | |
8e669ac1 | 1336 | # risking an integer overflow when generating the pie charts. |
8974000d SC |
1337 | if ($$m_data_gigs{$key}) { |
1338 | $use_gig = 1; | |
1339 | } | |
d5692f86 SC |
1340 | |
1341 | $rounded_volume = volume_rounded($$m_data{$key},$$m_data_gigs{$key}); | |
1342 | $data = $gigs = 0; | |
1343 | un_round($rounded_volume,\$data,\$gigs); | |
1344 | $count = $$m_count{$key}; | |
1345 | $rounded_average = volume_rounded($data/$count,$gigs/$count); | |
1346 | my @content=($count, $rounded_volume, $rounded_average ); | |
1347 | ||
1348 | # write content | |
1349 | # any reasons for not including rounded_average in the txt-version?? -fh | |
1350 | printf $txt_fh ($txt_format, $count, $rounded_volume, $key) if $txt_fh; | |
1351 | if ($htm_fh) { | |
059ec3d9 PH |
1352 | $htmlkey = $key; |
1353 | $htmlkey =~ s/>/\>\;/g; | |
1354 | $htmlkey =~ s/</\<\;/g; | |
d5692f86 | 1355 | printf $htm_fh ($htm_format, @content, $htmlkey); |
059ec3d9 | 1356 | } |
d5692f86 SC |
1357 | if ($xls_fh) |
1358 | { | |
1359 | &set_worksheet_line($ws_top50, $row_league_table++, 0, [@content, $key], $f_default); | |
059ec3d9 PH |
1360 | } |
1361 | ||
d5692f86 | 1362 | |
8974000d SC |
1363 | if (scalar @chartdatanames < $ntopchart) { |
1364 | if ($use_gig) { | |
1365 | if ($$m_data_gigs{$key}) { | |
d5692f86 SC |
1366 | push(@chartdatanames, $key); |
1367 | push(@chartdatavals, $$m_data_gigs{$key}); | |
8974000d | 1368 | } |
059ec3d9 | 1369 | } |
8974000d SC |
1370 | else { |
1371 | push(@chartdatanames, $key); | |
1372 | push(@chartdatavals, $$m_data{$key}); | |
059ec3d9 PH |
1373 | } |
1374 | } | |
8974000d SC |
1375 | else { |
1376 | $chartotherval += ($use_gig) ? $$m_data_gigs{$key} : $$m_data{$key}; | |
1377 | } | |
1378 | } | |
059ec3d9 PH |
1379 | push(@chartdatanames, "Other"); |
1380 | push(@chartdatavals, $chartotherval); | |
1381 | ||
d5692f86 SC |
1382 | print $txt_fh "\n" if $txt_fh; |
1383 | if ($htm_fh) { | |
1384 | print $htm_fh "</table>\n"; | |
1385 | print $htm_fh "</td><td>\n"; | |
059ec3d9 PH |
1386 | if ($HAVE_GD_Graph_pie && $charts) { |
1387 | # calculate the graph | |
1388 | my @data = ( | |
1389 | \@chartdatanames, | |
1390 | \@chartdatavals | |
1391 | ); | |
1392 | my $graph = GD::Graph::pie->new(300, 300); | |
1393 | $graph->set( | |
1394 | x_label => 'Name', | |
8974000d | 1395 | y_label => 'Volume' , |
059ec3d9 PH |
1396 | title => 'By Volume', |
1397 | ); | |
1398 | my $gd = $graph->plot(\@data) or warn($graph->error); | |
1399 | if ($gd) { | |
d5692f86 | 1400 | $temp = $text; |
059ec3d9 | 1401 | $temp =~ s/ /_/g; |
d5692f86 | 1402 | open(IMG, ">$chartdir/${temp}_volume.png") or die "Could not write $chartdir/${temp}_volume.png: $!\n"; |
059ec3d9 PH |
1403 | binmode IMG; |
1404 | print IMG $gd->png; | |
1405 | close IMG; | |
d5692f86 | 1406 | print $htm_fh "<img src=\"$chartrel/${temp}_volume.png\">"; |
059ec3d9 PH |
1407 | } |
1408 | } | |
d5692f86 SC |
1409 | print $htm_fh "</td><td>\n"; |
1410 | print $htm_fh "</td></tr></table>\n\n"; | |
059ec3d9 | 1411 | } |
d5692f86 SC |
1412 | if ($xls_fh) |
1413 | { | |
1414 | ++$row_league_table; | |
059ec3d9 PH |
1415 | } |
1416 | ||
d5692f86 | 1417 | } |
059ec3d9 | 1418 | |
1b4fe9dd PH |
1419 | ####################################################################### |
1420 | # top_n_sort(); | |
8e669ac1 | 1421 | # |
1b4fe9dd | 1422 | # @sorted_keys = top_n_sort($n,$href1,$href2,$href3); |
8e669ac1 | 1423 | # |
1b4fe9dd PH |
1424 | # Given a hash which has numerical values, return the sorted $n keys which |
1425 | # point to the top values. The second and third hashes are used as | |
1426 | # tiebreakers. They all must have the same keys. | |
8e669ac1 | 1427 | # |
1b4fe9dd PH |
1428 | # The idea behind this routine is that when you only want to see the |
1429 | # top n members of a set, rather than sorting the entire set and then | |
1430 | # plucking off the top n, sort through the stack as you go, discarding | |
1431 | # any member which is lower than your current n'th highest member. | |
8e669ac1 | 1432 | # |
1b4fe9dd PH |
1433 | # This proves to be an order of magnitude faster for large hashes. |
1434 | # On 200,000 lines of mainlog it benchmarked 9 times faster. | |
1435 | # On 700,000 lines of mainlog it benchmarked 13.8 times faster. | |
8e669ac1 | 1436 | # |
1b4fe9dd PH |
1437 | # We assume the values are > 0. |
1438 | ####################################################################### | |
059ec3d9 PH |
1439 | sub top_n_sort { |
1440 | my($n,$href1,$href2,$href3) = @_; | |
1441 | ||
1442 | # PH's original sort was: | |
1443 | # | |
1444 | # foreach $key (sort | |
1445 | # { | |
1446 | # $$m_count{$b} <=> $$m_count{$a} || | |
1447 | # $$m_data_gigs{$b} <=> $$m_data_gigs{$a} || | |
1448 | # $$m_data{$b} <=> $$m_data{$a} || | |
1449 | # $a cmp $b | |
1450 | # } | |
1451 | # keys %{$m_count}) | |
1452 | # | |
1453 | ||
1454 | #We use a key of '_' to represent non-existant values, as null keys are valid. | |
1455 | #'_' is not a valid domain, edomain, host, or email. | |
1456 | my(@top_n_keys) = ('_') x $n; | |
1457 | my($minimum_value1,$minimum_value2,$minimum_value3) = (0,0,0); | |
1458 | my $top_n_key = ''; | |
1459 | my $n_minus_1 = $n - 1; | |
1460 | my $n_minus_2 = $n - 2; | |
1461 | ||
8e669ac1 | 1462 | # Pick out the top $n keys. |
059ec3d9 PH |
1463 | my($key,$value1,$value2,$value3,$i,$comparison,$insert_position); |
1464 | while (($key,$value1) = each %$href1) { | |
1465 | ||
1466 | #print STDERR "key $key ($value1,",$href2->{$key},",",$href3->{$key},") <=> ($minimum_value1,$minimum_value2,$minimum_value3)\n"; | |
8e669ac1 | 1467 | |
059ec3d9 PH |
1468 | # Check to see that the new value is bigger than the lowest of the |
1469 | # top n keys that we're keeping. | |
8e669ac1 | 1470 | $comparison = $value1 <=> $minimum_value1 || |
d5692f86 SC |
1471 | $href2->{$key} <=> $minimum_value2 || |
1472 | $href3->{$key} <=> $minimum_value3 || | |
1473 | $top_n_key cmp $key; | |
059ec3d9 PH |
1474 | next unless ($comparison == 1); |
1475 | ||
1476 | # As we will be using these values a few times, extract them into scalars. | |
1477 | $value2 = $href2->{$key}; | |
1478 | $value3 = $href3->{$key}; | |
1479 | ||
1480 | # This key is bigger than the bottom n key, so the lowest position we | |
1481 | # will insert it into is $n minus 1 (the bottom of the list). | |
1482 | $insert_position = $n_minus_1; | |
1483 | ||
1484 | # Now go through the list, stopping when we find a key that we're | |
1485 | # bigger than, or we come to the penultimate position - we've | |
1486 | # already tested bigger than the last. | |
1487 | # | |
1488 | # Note: we go top down as the list starts off empty. | |
1489 | # Note: stepping through the list in this way benchmarks nearly | |
1490 | # three times faster than doing a sort() on the reduced list. | |
1491 | # I assume this is because the list is already in order, and | |
1492 | # we get a performance boost from not having to do hash lookups | |
1493 | # on the new key. | |
1494 | for ($i = 0; $i < $n_minus_1; $i++) { | |
1495 | $top_n_key = $top_n_keys[$i]; | |
1496 | if ( ($top_n_key eq '_') || | |
d5692f86 | 1497 | ( ($value1 <=> $href1->{$top_n_key} || |
059ec3d9 | 1498 | $value2 <=> $href2->{$top_n_key} || |
d5692f86 SC |
1499 | $value3 <=> $href3->{$top_n_key} || |
1500 | $top_n_key cmp $key) == 1 | |
1501 | ) | |
1502 | ) { | |
1503 | $insert_position = $i; | |
1504 | last; | |
059ec3d9 PH |
1505 | } |
1506 | } | |
1507 | ||
1508 | # Remove the last element, then insert the new one. | |
1509 | $#top_n_keys = $n_minus_2; | |
1510 | splice(@top_n_keys,$insert_position,0,$key); | |
1511 | ||
1512 | # Extract our new minimum values. | |
1513 | $top_n_key = $top_n_keys[$n_minus_1]; | |
1514 | if ($top_n_key ne '_') { | |
1515 | $minimum_value1 = $href1->{$top_n_key}; | |
1516 | $minimum_value2 = $href2->{$top_n_key}; | |
1517 | $minimum_value3 = $href3->{$top_n_key}; | |
1518 | } | |
1519 | } | |
1520 | ||
1521 | # Return the top n list, grepping out non-existant values, just in case | |
1522 | # we didn't have that many values. | |
1523 | return(grep(!/^_$/,@top_n_keys)); | |
1524 | } | |
1525 | ||
1526 | ||
1b4fe9dd PH |
1527 | ####################################################################### |
1528 | # html_header(); | |
8e669ac1 | 1529 | # |
1b4fe9dd | 1530 | # $header = html_header($title); |
8e669ac1 | 1531 | # |
1b4fe9dd PH |
1532 | # Print our HTML header and start the <body> block. |
1533 | ####################################################################### | |
059ec3d9 PH |
1534 | sub html_header { |
1535 | my($title) = @_; | |
1536 | my $text = << "EoText"; | |
1537 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"> | |
1538 | <html> | |
1539 | <head> | |
1540 | <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-15"> | |
1541 | <title>$title</title> | |
1542 | </head> | |
1543 | <body bgcolor="white"> | |
1544 | <h1>$title</h1> | |
1545 | EoText | |
1546 | return $text; | |
1547 | } | |
1548 | ||
1549 | ||
1550 | ||
1b4fe9dd PH |
1551 | ####################################################################### |
1552 | # help(); | |
8e669ac1 | 1553 | # |
1b4fe9dd | 1554 | # help(); |
8e669ac1 | 1555 | # |
1b4fe9dd PH |
1556 | # Display usage instructions and exit. |
1557 | ####################################################################### | |
059ec3d9 PH |
1558 | sub help { |
1559 | print << "EoText"; | |
1560 | ||
1561 | eximstats Version $VERSION | |
1562 | ||
d5692f86 SC |
1563 | Usage: |
1564 | eximstats [Output] [Options] mainlog1 mainlog2 ... | |
1565 | eximstats -merge -html [Options] report.1.html ... > weekly_rep.html | |
1566 | ||
1567 | Examples: | |
1568 | eximstats -html=eximstats.html mainlog1 mainlog2 ... | |
1569 | eximstats mainlog1 mainlog2 ... > report.txt | |
059ec3d9 | 1570 | |
8974000d | 1571 | Parses exim mainlog or syslog files and generates a statistical analysis |
d5692f86 SC |
1572 | of the messages processed. |
1573 | ||
1574 | Valid output types are: | |
1575 | -txt[=<file>] plain text (default unless no other type is specified) | |
1576 | -html[=<file>] HTML | |
1577 | -xls[=<file>] Excel | |
1578 | With no type and file given, defaults to -txt and STDOUT. | |
059ec3d9 | 1579 | |
d5692f86 | 1580 | Valid options are: |
059ec3d9 PH |
1581 | -h<number> histogram divisions per hour. The default is 1, and |
1582 | 0 suppresses histograms. Other valid values are: | |
d5692f86 | 1583 | 2, 3, 5, 10, 15, 20, 30 or 60. |
059ec3d9 PH |
1584 | -ne don't display error information |
1585 | -nr don't display relaying information | |
1586 | -nr/pattern/ don't display relaying information that matches | |
1587 | -nt don't display transport information | |
1588 | -nt/pattern/ don't display transport information that matches | |
d5692f86 | 1589 | -nvr don't do volume rounding. Display in bytes, not KB/MB/GB. |
059ec3d9 PH |
1590 | -q<list> list of times for queuing information |
1591 | single 0 item suppresses | |
1592 | -t<number> display top <number> sources/destinations | |
1593 | default is 50, 0 suppresses top listing | |
1594 | -tnl omit local sources/destinations in top listing | |
1595 | -t_remote_users show top user sources/destinations from non-local domains | |
d5692f86 SC |
1596 | -include_original_destination show both the final and original |
1597 | destinations in the results rather than just the final ones. | |
059ec3d9 | 1598 | |
d5692f86 | 1599 | -byhost show results by sending host (default unless bydomain or |
059ec3d9 | 1600 | byemail is specified) |
d5692f86 SC |
1601 | -bydomain show results by sending domain. |
1602 | -byemail show results by sender's email address | |
1603 | -byedomain show results by sender's email domain | |
059ec3d9 PH |
1604 | |
1605 | -pattern "Description" /pattern/ | |
1606 | Count lines matching specified patterns and show them in | |
d5692f86 SC |
1607 | the results. It can be specified multiple times. Eg: |
1608 | -pattern 'Refused connections' '/refused connection/' | |
059ec3d9 PH |
1609 | |
1610 | -merge merge previously generated reports into a new report | |
1611 | ||
d5692f86 SC |
1612 | -charts Create charts (this requires the GD::Graph modules). |
1613 | Only valid with -html. | |
059ec3d9 PH |
1614 | -chartdir <dir> Create the charts' png files in the directory <dir> |
1615 | -chartrel <dir> Specify the relative directory for the "img src=" tags | |
1616 | from where to include the charts in the html file | |
d5692f86 | 1617 | -chartdir and -chartrel default to '.' |
059ec3d9 | 1618 | |
d5692f86 | 1619 | -d Debug mode - dump the eval'ed parser onto STDERR. |
059ec3d9 PH |
1620 | |
1621 | EoText | |
1622 | ||
1623 | exit 1; | |
1624 | } | |
1625 | ||
1626 | ||
1627 | ||
1b4fe9dd PH |
1628 | ####################################################################### |
1629 | # generate_parser(); | |
8e669ac1 | 1630 | # |
1b4fe9dd | 1631 | # $parser = generate_parser(); |
8e669ac1 | 1632 | # |
1b4fe9dd PH |
1633 | # This subroutine generates the parsing routine which will be |
1634 | # used to parse the mainlog. We take the base operation, and remove bits not in use. | |
1635 | # This improves performance depending on what bits you take out or add. | |
8e669ac1 | 1636 | # |
1b4fe9dd | 1637 | # I've tested using study(), but this does not improve performance. |
8e669ac1 | 1638 | # |
1b4fe9dd PH |
1639 | # We store our parsing routing in a variable, and process it looking for #IFDEF (Expression) |
1640 | # or #IFNDEF (Expression) statements and corresponding #ENDIF (Expression) statements. If | |
1641 | # the expression evaluates to true, then it is included/excluded accordingly. | |
1642 | ####################################################################### | |
059ec3d9 PH |
1643 | sub generate_parser { |
1644 | my $parser = ' | |
1645 | my($ip,$host,$email,$edomain,$domain,$thissize,$size,$old,$new); | |
1646 | my($tod,$m_hour,$m_min,$id,$flag); | |
1647 | while (<$fh>) { | |
059ec3d9 | 1648 | |
8974000d SC |
1649 | # Convert syslog lines to mainlog format. |
1650 | if (! /^\\d{4}/) { | |
1651 | next unless s/^.*? exim\\b.*?: //; | |
1652 | } | |
1653 | ||
1654 | next if length($_) < 38; | |
059ec3d9 PH |
1655 | next unless /^(\\d{4}\\-\\d\\d-\\d\\d\\s(\\d\\d):(\\d\\d):\\d\\d( [-+]\\d\\d\\d\\d)?)/o; |
1656 | ||
1657 | ($tod,$m_hour,$m_min) = ($1,$2,$3); | |
1658 | ||
1659 | # PH | |
1660 | my($extra) = defined($4)? 6 : 0; | |
1661 | $id = substr($_, 20 + $extra, 16); | |
1662 | $flag = substr($_, 37 + $extra, 2); | |
1663 | '; | |
1664 | ||
1665 | # Watch for user specified patterns. | |
1666 | my $user_pattern_index = 0; | |
1667 | foreach (@user_patterns) { | |
1668 | $user_pattern_totals[$user_pattern_index] = 0; | |
1669 | $parser .= " \$user_pattern_totals[$user_pattern_index]++ if $_;\n"; | |
1670 | $user_pattern_index++; | |
1671 | } | |
1672 | ||
1673 | $parser .= ' | |
1674 | next unless ($flag =~ /<=|=>|->|==|\\*\\*|Co/); | |
1675 | ||
1676 | #Strip away the timestamp, ID and flag (which could be "Com" for completed) | |
1677 | #This speeds up the later pattern matches. | |
1678 | # $_ = substr($_, 40); | |
1679 | ||
1680 | $_ = substr($_, 40 + $extra); # PH | |
1681 | ||
1682 | # JN - Skip over certain transports as specified via the "-nt/.../" command | |
1683 | # line switch (where ... is a perl style regular expression). This is | |
1684 | # required so that transports that skew stats such as SpamAssassin can be | |
1685 | # ignored. | |
1686 | #IFDEF ($transport_pattern) | |
1687 | if (/\\sT=(\\S+)/) { | |
1688 | next if ($1 =~ /$transport_pattern/o) ; | |
1689 | } | |
1690 | #ENDIF ($transport_pattern) | |
1691 | ||
1692 | ||
d5692f86 SC |
1693 | $host = "local"; #Host is local unless otherwise specified. |
1694 | $domain = "localdomain"; #Domain is localdomain unless otherwise specified. | |
059ec3d9 PH |
1695 | |
1696 | ||
1697 | # Do some pattern matches to get the host and IP address. | |
1698 | # We expect lines to be of the form "H=[IpAddr]" or "H=Host [IpAddr]" or | |
1699 | # "H=Host (UnverifiedHost) [IpAddr]" or "H=(UnverifiedHost) [IpAddr]". | |
1700 | # We do 2 separate matches to keep the matches simple and fast. | |
1701 | if (/\\sH=(\\S+)/) { | |
1702 | $host = $1; | |
1703 | ||
1704 | ($ip) = /\\sH=.*?(\\s\\[[^]]+\\])/; | |
1705 | # If there is only an IP address, it will be in $host and $ip will be | |
1706 | # unset. That is OK, because we only use $ip in conjunction with $host | |
1707 | # below. But make it empty to avoid warning messages. | |
1708 | $ip = "" if !defined $ip; | |
1709 | ||
1710 | #IFDEF ($do_sender{Domain}) | |
1711 | if ($host !~ /^\\[/ && $host =~ /^(\\(?)[^\\.]+\\.([^\\.]+\\..*)/) { | |
d5692f86 SC |
1712 | # Remove the host portion from the DNS name. We ensure that we end up |
1713 | # with at least xxx.yyy. $host can be "(x.y.z)" or "x.y.z". | |
1714 | $domain = lc("$1.$2"); | |
1715 | $domain =~ s/^\\.//; #Remove preceding dot. | |
059ec3d9 PH |
1716 | } |
1717 | #ENDIF ($do_sender{Domain}) | |
1718 | ||
1719 | } | |
1720 | ||
1721 | #IFDEF ($do_sender{Email}) | |
d5692f86 SC |
1722 | #IFDEF ($include_original_destination) |
1723 | # Catch both "a@b.com <c@d.com>" and "e@f.com" | |
1724 | #$email = (/^(\S+) (<(\S*?)>)?/) ? $3 || $1 : ""; | |
1725 | $email = (/^(\S+ (<[^@>]+@?[^>]*>)?)/) ? $1 : ""; | |
1726 | chomp($email); | |
1727 | #ENDIF ($include_original_destination) | |
1728 | ||
1729 | #IFNDEF ($include_original_destination) | |
1730 | $email = (/^(\S+)/) ? $1 : ""; | |
1731 | #ENDIF ($include_original_destination) | |
059ec3d9 PH |
1732 | #ENDIF ($do_sender{Email}) |
1733 | ||
1734 | #IFDEF ($do_sender{Edomain}) | |
d5692f86 SC |
1735 | #IFDEF ($include_original_destination) |
1736 | #$edomain = (/^(\S+) (<\S*?\\@(\S+)>)?/) ? $3 || $1 : ""; | |
1737 | $edomain = (/^(\S+ (<\S*?\\@(\S+?)>)?)/) ? $1 : ""; | |
1738 | chomp($edomain); | |
1739 | lc($edomain); | |
1740 | #ENDIF ($include_original_destination) | |
1741 | ||
1742 | #IFNDEF ($include_original_destination) | |
1743 | $edomain = (/^\S*?\\@(\S+)/) ? lc($1) : ""; | |
1744 | #ENDIF ($include_original_destination) | |
059ec3d9 PH |
1745 | #ENDIF ($do_sender{Edomain}) |
1746 | ||
1747 | if ($tod lt $begin) { | |
1748 | $begin = $tod; | |
1749 | } | |
1750 | elsif ($tod gt $end) { | |
1751 | $end = $tod; | |
1752 | } | |
1753 | ||
1754 | ||
1755 | if ($flag eq "<=") { | |
1756 | $thissize = (/\\sS=(\\d+)( |$)/) ? $1 : 0; | |
1757 | $size{$id} = $thissize; | |
1758 | ||
1759 | #IFDEF ($show_relay) | |
1760 | if ($host ne "local") { | |
d5692f86 SC |
1761 | # Save incoming information in case it becomes interesting |
1762 | # later, when delivery lines are read. | |
1763 | my($from) = /^(\\S+)/; | |
1764 | $from_host{$id} = "$host$ip"; | |
1765 | $from_address{$id} = $from; | |
059ec3d9 PH |
1766 | } |
1767 | #ENDIF ($show_relay) | |
1768 | ||
1769 | #IFDEF ($local_league_table || $include_remote_users) | |
d5692f86 SC |
1770 | if (/\sU=(\\S+)/) { |
1771 | my $user = $1; | |
059ec3d9 | 1772 | |
d5692f86 SC |
1773 | #IFDEF ($local_league_table && $include_remote_users) |
1774 | { #Store both local and remote users. | |
1775 | #ENDIF ($local_league_table && $include_remote_users) | |
059ec3d9 | 1776 | |
d5692f86 SC |
1777 | #IFDEF ($local_league_table && ! $include_remote_users) |
1778 | if ($host eq "local") { #Store local users only. | |
1779 | #ENDIF ($local_league_table && ! $include_remote_users) | |
059ec3d9 | 1780 | |
d5692f86 SC |
1781 | #IFDEF ($include_remote_users && ! $local_league_table) |
1782 | if ($host ne "local") { #Store remote users only. | |
1783 | #ENDIF ($include_remote_users && ! $local_league_table) | |
059ec3d9 | 1784 | |
d5692f86 SC |
1785 | $received_count_user{$user}++; |
1786 | add_volume(\\$received_data_user{$user},\\$received_data_gigs_user{$user},$thissize); | |
059ec3d9 | 1787 | } |
d5692f86 | 1788 | } |
059ec3d9 PH |
1789 | #ENDIF ($local_league_table || $include_remote_users) |
1790 | ||
1791 | #IFDEF ($do_sender{Host}) | |
d5692f86 SC |
1792 | $received_count{Host}{$host}++; |
1793 | add_volume(\\$received_data{Host}{$host},\\$received_data_gigs{Host}{$host},$thissize); | |
059ec3d9 PH |
1794 | #ENDIF ($do_sender{Host}) |
1795 | ||
1796 | #IFDEF ($do_sender{Domain}) | |
1797 | if ($domain) { | |
d5692f86 SC |
1798 | $received_count{Domain}{$domain}++; |
1799 | add_volume(\\$received_data{Domain}{$domain},\\$received_data_gigs{Domain}{$domain},$thissize); | |
1800 | } | |
059ec3d9 PH |
1801 | #ENDIF ($do_sender{Domain}) |
1802 | ||
1803 | #IFDEF ($do_sender{Email}) | |
d5692f86 SC |
1804 | $received_count{Email}{$email}++; |
1805 | add_volume(\\$received_data{Email}{$email},\\$received_data_gigs{Email}{$email},$thissize); | |
059ec3d9 PH |
1806 | #ENDIF ($do_sender{Email}) |
1807 | ||
1808 | #IFDEF ($do_sender{Edomain}) | |
d5692f86 SC |
1809 | $received_count{Edomain}{$edomain}++; |
1810 | add_volume(\\$received_data{Edomain}{$edomain},\\$received_data_gigs{Edomain}{$edomain},$thissize); | |
059ec3d9 PH |
1811 | #ENDIF ($do_sender{Edomain}) |
1812 | ||
1813 | $total_received_count++; | |
1814 | add_volume(\\$total_received_data,\\$total_received_data_gigs,$thissize); | |
1815 | ||
1816 | #IFDEF ($#queue_times >= 0) | |
d5692f86 | 1817 | $arrival_time{$id} = $tod; |
059ec3d9 PH |
1818 | #ENDIF ($#queue_times >= 0) |
1819 | ||
1820 | #IFDEF ($hist_opt > 0) | |
d5692f86 | 1821 | $received_interval_count[($m_hour*60 + $m_min)/$hist_interval]++; |
059ec3d9 PH |
1822 | #ENDIF ($hist_opt > 0) |
1823 | } | |
1824 | ||
1825 | elsif ($flag eq "=>") { | |
1826 | $size = $size{$id} || 0; | |
1827 | if ($host ne "local") { | |
1828 | $remote_delivered{$id} = 1; | |
1829 | ||
1830 | ||
1831 | #IFDEF ($show_relay) | |
1832 | # Determine relaying address if either only one address listed, | |
1833 | # or two the same. If they are different, it implies a forwarding | |
1834 | # or aliasing, which is not relaying. Note that for multi-aliased | |
1835 | # addresses, there may be a further address between the first | |
1836 | # and last. | |
1837 | ||
1838 | if (defined $from_host{$id}) { | |
1839 | if (/^(\\S+)(?:\\s+\\([^)]\\))?\\s+<([^>]+)>/) { | |
1840 | ($old,$new) = ($1,$2); | |
d5692f86 | 1841 | } |
059ec3d9 | 1842 | else { |
d5692f86 SC |
1843 | $old = $new = ""; |
1844 | } | |
059ec3d9 PH |
1845 | |
1846 | if ("\\L$new" eq "\\L$old") { | |
1847 | ($old) = /^(\\S+)/ if $old eq ""; | |
1848 | my $key = "H=\\L$from_host{$id}\\E A=\\L$from_address{$id}\\E => " . | |
1849 | "H=\\L$host\\E$ip A=\\L$old\\E"; | |
1850 | if (!defined $relay_pattern || $key !~ /$relay_pattern/o) { | |
1851 | $relayed{$key} = 0 if !defined $relayed{$key}; | |
1852 | $relayed{$key}++; | |
d5692f86 | 1853 | } |
059ec3d9 | 1854 | else { |
d5692f86 | 1855 | $relayed_unshown++ |
059ec3d9 PH |
1856 | } |
1857 | } | |
1858 | } | |
1859 | #ENDIF ($show_relay) | |
1860 | ||
1861 | } | |
1862 | ||
1863 | #IFDEF ($local_league_table || $include_remote_users) | |
d5692f86 SC |
1864 | #IFDEF ($local_league_table && $include_remote_users) |
1865 | { #Store both local and remote users. | |
1866 | #ENDIF ($local_league_table && $include_remote_users) | |
1867 | ||
1868 | #IFDEF ($local_league_table && ! $include_remote_users) | |
1869 | if ($host eq "local") { #Store local users only. | |
1870 | #ENDIF ($local_league_table && ! $include_remote_users) | |
1871 | ||
1872 | #IFDEF ($include_remote_users && ! $local_league_table) | |
1873 | if ($host ne "local") { #Store remote users only. | |
1874 | #ENDIF ($include_remote_users && ! $local_league_table) | |
1875 | ||
1876 | if (my($user) = split((/\\s</)? " <" : " ", $_)) { | |
1877 | #IFDEF ($include_original_destination) | |
1878 | { | |
1879 | #ENDIF ($include_original_destination) | |
1880 | #IFNDEF ($include_original_destination) | |
1881 | if ($user =~ /^[\\/|]/) { | |
1882 | #ENDIF ($include_original_destination) | |
1883 | my($parent) = $_ =~ /(<[^@]+@?[^>]*>)/; | |
1884 | $user = "$user $parent" if defined $parent; | |
1885 | } | |
1886 | $delivered_count_user{$user}++; | |
1887 | add_volume(\\$delivered_data_user{$user},\\$delivered_data_gigs_user{$user},$size); | |
1888 | } | |
1889 | } | |
059ec3d9 PH |
1890 | #ENDIF ($local_league_table || $include_remote_users) |
1891 | ||
1892 | #IFDEF ($do_sender{Host}) | |
d5692f86 SC |
1893 | $delivered_count{Host}{$host}++; |
1894 | add_volume(\\$delivered_data{Host}{$host},\\$delivered_data_gigs{Host}{$host},$size); | |
059ec3d9 PH |
1895 | #ENDIF ($do_sender{Host}) |
1896 | #IFDEF ($do_sender{Domain}) | |
1897 | if ($domain) { | |
d5692f86 SC |
1898 | $delivered_count{Domain}{$domain}++; |
1899 | add_volume(\\$delivered_data{Domain}{$domain},\\$delivered_data_gigs{Domain}{$domain},$size); | |
1900 | } | |
059ec3d9 PH |
1901 | #ENDIF ($do_sender{Domain}) |
1902 | #IFDEF ($do_sender{Email}) | |
d5692f86 SC |
1903 | $delivered_count{Email}{$email}++; |
1904 | add_volume(\\$delivered_data{Email}{$email},\\$delivered_data_gigs{Email}{$email},$size); | |
059ec3d9 PH |
1905 | #ENDIF ($do_sender{Email}) |
1906 | #IFDEF ($do_sender{Edomain}) | |
d5692f86 SC |
1907 | $delivered_count{Edomain}{$edomain}++; |
1908 | add_volume(\\$delivered_data{Edomain}{$edomain},\\$delivered_data_gigs{Edomain}{$edomain},$size); | |
059ec3d9 PH |
1909 | #ENDIF ($do_sender{Edomain}) |
1910 | ||
1911 | $total_delivered_count++; | |
1912 | add_volume(\\$total_delivered_data,\\$total_delivered_data_gigs,$size); | |
1913 | ||
1914 | #IFDEF ($show_transport) | |
1915 | my $transport = (/\\sT=(\\S+)/) ? $1 : ":blackhole:"; | |
1916 | $transported_count{$transport}++; | |
1917 | add_volume(\\$transported_data{$transport},\\$transported_data_gigs{$transport},$size); | |
1918 | #ENDIF ($show_transport) | |
1919 | ||
1920 | #IFDEF ($hist_opt > 0) | |
1921 | $delivered_interval_count[($m_hour*60 + $m_min)/$hist_interval]++; | |
1922 | #ENDIF ($hist_opt > 0) | |
1923 | ||
1924 | } | |
1925 | ||
1926 | elsif ($flag eq "==" && defined($size{$id}) && !defined($delayed{$id})) { | |
1927 | $delayed_count++; | |
1928 | $delayed{$id} = 1; | |
1929 | } | |
1930 | ||
1931 | elsif ($flag eq "**") { | |
1932 | $had_error{$id} = 1 if defined ($size{$id}); | |
1933 | ||
1934 | #IFDEF ($show_errors) | |
1935 | $errors_count{$_}++; | |
1936 | #ENDIF ($show_errors) | |
1937 | ||
1938 | } | |
1939 | ||
1940 | elsif ($flag eq "Co") { | |
1941 | #Completed? | |
1942 | #IFDEF ($#queue_times >= 0) | |
1943 | #Note: id_seconds() benchmarks as 42% slower than seconds() and computing | |
1944 | #the time accounts for a significant portion of the run time. | |
1945 | my($queued); | |
1946 | if (defined $arrival_time{$id}) { | |
1947 | $queued = seconds($tod) - seconds($arrival_time{$id}); | |
d5692f86 | 1948 | delete($arrival_time{$id}); |
059ec3d9 PH |
1949 | } |
1950 | else { | |
d5692f86 | 1951 | $queued = seconds($tod) - id_seconds($id); |
059ec3d9 PH |
1952 | } |
1953 | ||
1954 | for ($i = 0; $i <= $#queue_times; $i++) { | |
1955 | if ($queued < $queue_times[$i]) { | |
1956 | $queue_bin[$i]++; | |
1957 | $remote_queue_bin[$i]++ if $remote_delivered{$id}; | |
1958 | last; | |
d5692f86 SC |
1959 | } |
1960 | } | |
059ec3d9 PH |
1961 | $queue_more_than++ if $i > $#queue_times; |
1962 | #ENDIF ($#queue_times >= 0) | |
1963 | ||
1964 | #IFDEF ($show_relay) | |
1965 | delete($from_host{$id}); | |
1966 | delete($from_address{$id}); | |
1967 | #ENDIF ($show_relay) | |
1968 | ||
1969 | } | |
1970 | }'; | |
1971 | ||
1972 | # We now do a 'C preprocessor style operation on our parser | |
1973 | # to remove bits not in use. | |
1974 | my(%defines_in_operation,$removing_lines,$processed_parser); | |
1975 | foreach (split (/\n/,$parser)) { | |
1976 | if ((/^\s*#\s*IFDEF\s*\((.*?)\)/i && ! eval $1) || | |
d5692f86 | 1977 | (/^\s*#\s*IFNDEF\s*\((.*?)\)/i && eval $1) ) { |
059ec3d9 PH |
1978 | $defines_in_operation{$1} = 1; |
1979 | $removing_lines = 1; | |
1980 | } | |
1981 | ||
1982 | $processed_parser .= $_."\n" unless $removing_lines; | |
1983 | ||
1984 | if (/^\s*#\s*ENDIF\s*\((.*?)\)/i) { | |
1985 | delete $defines_in_operation{$1}; | |
1986 | unless (keys %defines_in_operation) { | |
d5692f86 | 1987 | $removing_lines = 0; |
059ec3d9 PH |
1988 | } |
1989 | } | |
1990 | } | |
1991 | print STDERR "# START OF PARSER:\n$processed_parser\n# END OF PARSER\n\n" if $debug; | |
1992 | ||
1993 | return $processed_parser; | |
1994 | } | |
1995 | ||
1996 | ||
1997 | ||
1b4fe9dd PH |
1998 | ####################################################################### |
1999 | # parse(); | |
8e669ac1 | 2000 | # |
1b4fe9dd | 2001 | # parse($parser,\*FILEHANDLE); |
8e669ac1 | 2002 | # |
1b4fe9dd PH |
2003 | # This subroutine accepts a parser and a filehandle from main and parses each |
2004 | # line. We store the results into global variables. | |
2005 | ####################################################################### | |
059ec3d9 PH |
2006 | sub parse { |
2007 | my($parser,$fh) = @_; | |
2008 | ||
2009 | if ($merge_reports) { | |
2010 | parse_old_eximstat_reports($fh); | |
2011 | } | |
2012 | else { | |
2013 | eval $parser; | |
2014 | die ($@) if $@; | |
2015 | } | |
2016 | ||
2017 | } | |
2018 | ||
2019 | ||
2020 | ||
1b4fe9dd PH |
2021 | ####################################################################### |
2022 | # print_header(); | |
8e669ac1 | 2023 | # |
1b4fe9dd | 2024 | # print_header(); |
8e669ac1 | 2025 | # |
1b4fe9dd PH |
2026 | # Print our headers and contents. |
2027 | ####################################################################### | |
059ec3d9 PH |
2028 | sub print_header { |
2029 | ||
d5692f86 | 2030 | |
059ec3d9 PH |
2031 | my $title = "Exim statistics from $begin to $end"; |
2032 | ||
d5692f86 SC |
2033 | print $txt_fh "\n$title\n" if $txt_fh; |
2034 | if ($htm_fh) { | |
2035 | print $htm_fh html_header($title); | |
2036 | print $htm_fh "<ul>\n"; | |
2037 | print $htm_fh "<li><a href=\"#grandtotal\">Grand total summary</a>\n"; | |
2038 | print $htm_fh "<li><a href=\"#patterns\">User Specified Patterns</a>\n" if @user_patterns; | |
2039 | print $htm_fh "<li><a href=\"#transport\">Deliveries by Transport</a>\n" if $show_transport; | |
059ec3d9 | 2040 | if ($hist_opt) { |
d5692f86 SC |
2041 | print $htm_fh "<li><a href=\"#Messages received\">Messages received per hour</a>\n"; |
2042 | print $htm_fh "<li><a href=\"#Deliveries\">Deliveries per hour</a>\n"; | |
059ec3d9 PH |
2043 | } |
2044 | if ($#queue_times >= 0) { | |
d5692f86 SC |
2045 | print $htm_fh "<li><a href=\"#all messages time\">Time spent on the queue: all messages</a>\n"; |
2046 | print $htm_fh "<li><a href=\"#messages with at least one remote delivery time\">Time spent on the queue: messages with at least one remote delivery</a>\n"; | |
059ec3d9 | 2047 | } |
d5692f86 | 2048 | print $htm_fh "<li><a href=\"#Relayed messages\">Relayed messages</a>\n" if $show_relay; |
059ec3d9 PH |
2049 | if ($topcount) { |
2050 | foreach ('Host','Domain','Email','Edomain') { | |
d5692f86 SC |
2051 | next unless $do_sender{$_}; |
2052 | print $htm_fh "<li><a href=\"#sending \l$_ count\">Top $topcount sending \l${_}s by message count</a>\n"; | |
2053 | print $htm_fh "<li><a href=\"#sending \l$_ volume\">Top $topcount sending \l${_}s by volume</a>\n"; | |
059ec3d9 PH |
2054 | } |
2055 | if ($local_league_table || $include_remote_users) { | |
d5692f86 SC |
2056 | print $htm_fh "<li><a href=\"#local sender count\">Top $topcount local senders by message count</a>\n"; |
2057 | print $htm_fh "<li><a href=\"#local sender volume\">Top $topcount local senders by volume</a>\n"; | |
059ec3d9 PH |
2058 | } |
2059 | foreach ('Host','Domain','Email','Edomain') { | |
d5692f86 SC |
2060 | next unless $do_sender{$_}; |
2061 | print $htm_fh "<li><a href=\"#\l$_ destination count\">Top $topcount \l$_ destinations by message count</a>\n"; | |
2062 | print $htm_fh "<li><a href=\"#\l$_ destination volume\">Top $topcount \l$_ destinations by volume</a>\n"; | |
059ec3d9 PH |
2063 | } |
2064 | if ($local_league_table || $include_remote_users) { | |
d5692f86 SC |
2065 | print $htm_fh "<li><a href=\"#local destination count\">Top $topcount local destinations by message count</a>\n"; |
2066 | print $htm_fh "<li><a href=\"#local destination volume\">Top $topcount local destinations by volume</a>\n"; | |
059ec3d9 PH |
2067 | } |
2068 | } | |
d5692f86 SC |
2069 | print $htm_fh "<li><a href=\"#errors\">List of errors</a>\n" if %errors_count; |
2070 | print $htm_fh "</ul>\n<hr>\n"; | |
059ec3d9 | 2071 | } |
d5692f86 SC |
2072 | if ($xls_fh) |
2073 | { | |
2074 | $ws_global->write($row++, $col+0, "Exim Statistics", $f_header1); | |
2075 | &set_worksheet_line($ws_global, $row, $col, ["from:", $begin, "to:", $end], $f_default); | |
2076 | $row+=2; | |
059ec3d9 PH |
2077 | } |
2078 | } | |
2079 | ||
2080 | ||
1b4fe9dd PH |
2081 | ####################################################################### |
2082 | # print_grandtotals(); | |
8e669ac1 | 2083 | # |
1b4fe9dd | 2084 | # print_grandtotals(); |
8e669ac1 | 2085 | # |
1b4fe9dd PH |
2086 | # Print the grand totals. |
2087 | ####################################################################### | |
059ec3d9 PH |
2088 | sub print_grandtotals { |
2089 | ||
2090 | # Get the sender by headings and results. This is complicated as we can have | |
2091 | # different numbers of columns. | |
2092 | my($sender_txt_header,$sender_html_header,$sender_txt_format,$sender_html_format); | |
2093 | my(@received_totals,@delivered_totals); | |
d5692f86 SC |
2094 | my($row_tablehead, $row_max); |
2095 | ||
059ec3d9 PH |
2096 | foreach ('Host','Domain','Email','Edomain') { |
2097 | next unless $do_sender{$_}; | |
2098 | if ($merge_reports) { | |
2099 | push(@received_totals, get_report_total($report_totals{Received},"${_}s")); | |
2100 | push(@delivered_totals,get_report_total($report_totals{Delivered},"${_}s")); | |
2101 | } | |
2102 | else { | |
2103 | push(@received_totals,scalar(keys %{$received_data{$_}})); | |
2104 | push(@delivered_totals,scalar(keys %{$delivered_data{$_}})); | |
2105 | } | |
2106 | $sender_html_header .= "<th>${_}s</th>"; | |
2107 | $sender_txt_header .= " " x ($COLUMN_WIDTHS - length($_)) . $_ . 's'; | |
2108 | $sender_html_format .= "<td align=\"right\">%d</td>"; | |
2109 | $sender_txt_format .= " " x ($COLUMN_WIDTHS - 5) . "%6d"; | |
2110 | } | |
2111 | ||
d5692f86 SC |
2112 | my $txt_format1 = " %-16s %9s %6d $sender_txt_format"; |
2113 | my $txt_format2 = " %6d %4.1f%% %6d %4.1f%%", | |
2114 | my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%s</td>$sender_html_format<td align=\"right\">%d</td>"; | |
2115 | my $htm_format2 = "<td align=\"right\">%d</td><td align=\"right\">%4.1f%%</td><td align=\"right\">%d</td><td align=\"right\">%4.1f%%</td>"; | |
059ec3d9 | 2116 | |
d5692f86 | 2117 | if ($txt_fh) { |
059ec3d9 | 2118 | my $sender_spaces = " " x length($sender_txt_header); |
d5692f86 SC |
2119 | print $txt_fh "\n"; |
2120 | print $txt_fh "Grand total summary\n"; | |
2121 | print $txt_fh "-------------------\n"; | |
2122 | print $txt_fh " $sender_spaces At least one address\n"; | |
2123 | print $txt_fh " TOTAL Volume Messages $sender_txt_header Delayed Failed\n"; | |
2124 | } | |
2125 | if ($htm_fh) { | |
2126 | print $htm_fh "<a name=\"grandtotal\"></a>\n"; | |
2127 | print $htm_fh "<h2>Grand total summary</h2>\n"; | |
2128 | print $htm_fh "<table border=1>\n"; | |
2129 | print $htm_fh "<tr><th>TOTAL</th><th>Volume</th><th>Messages</th>$sender_html_header<th colspan=2>At least one addr<br>Delayed</th><th colspan=2>At least one addr<br>Failed</th>\n"; | |
2130 | } | |
2131 | if ($xls_fh) | |
2132 | { | |
2133 | $ws_global->write($row++, $col, "Grand total summary", $f_header2); | |
059ec3d9 | 2134 | |
d5692f86 SC |
2135 | $row_tablehead = $row+1; # header-row of TOTALS table |
2136 | ||
2137 | &set_worksheet_line($ws_global, $row_tablehead, 0, ['Received', 'Delivered', 'TOTAL'], $f_headertab); | |
2138 | ||
2139 | my @content= ( | |
2140 | "Volume", | |
2141 | "Messages", | |
2142 | $sender_txt_header, | |
2143 | "At least one address Delayed (Total)", | |
2144 | "At least one address Delayed (Percent)", | |
2145 | "At least one address Failed (Total)", | |
2146 | "At least one address Failed (Percent)" | |
2147 | ); | |
2148 | ||
2149 | for (my $i=0; $i < scalar(@content); $i++) | |
2150 | { | |
2151 | $ws_global->write($row_tablehead+$i+1, 2, $content[$i], $f_default); | |
2152 | $row++; | |
2153 | } | |
2154 | $row_max = $row_tablehead+scalar(@content)+2; # continue from this row | |
059ec3d9 PH |
2155 | } |
2156 | ||
d5692f86 SC |
2157 | |
2158 | ||
059ec3d9 PH |
2159 | my($volume,$failed_count); |
2160 | if ($merge_reports) { | |
2161 | $volume = volume_rounded($report_totals{Received}{Volume}, $report_totals{Received}{'Volume-gigs'}); | |
2162 | $total_received_count = get_report_total($report_totals{Received},'Messages'); | |
2163 | $failed_count = get_report_total($report_totals{Received},'Failed'); | |
2164 | $delayed_count = get_report_total($report_totals{Received},'Delayed'); | |
2165 | } | |
2166 | else { | |
2167 | $volume = volume_rounded($total_received_data, $total_received_data_gigs); | |
2168 | $failed_count = keys %had_error; | |
2169 | } | |
2170 | ||
2171 | { | |
2172 | no integer; | |
059ec3d9 | 2173 | |
d5692f86 SC |
2174 | my @content=( |
2175 | $volume,$total_received_count, | |
2176 | @received_totals, | |
2177 | $delayed_count, | |
2178 | ($total_received_count) ? ($delayed_count*100/$total_received_count) : 0, | |
2179 | $failed_count, | |
2180 | ($total_received_count) ? ($failed_count*100/$total_received_count) : 0 | |
2181 | ); | |
2182 | ||
2183 | printf $txt_fh ("$txt_format1$txt_format2\n", 'Received', @content) if $txt_fh; | |
2184 | printf $htm_fh ("$htm_format1$htm_format2\n", 'Received', @content) if $htm_fh; | |
2185 | if ($xls_fh) | |
2186 | { | |
2187 | $row = $row_tablehead+1; | |
2188 | for (my $i=0; $i < scalar(@content); $i++) | |
2189 | { | |
2190 | if ($i == 4 || $i == 6) { | |
2191 | $ws_global->write($row+$i, 0, $content[$i]/100, $f_percent); | |
2192 | } | |
2193 | else { | |
2194 | $ws_global->write($row+$i, 0, $content[$i], $f_default); | |
2195 | } | |
2196 | } | |
2197 | } | |
2198 | } | |
059ec3d9 PH |
2199 | if ($merge_reports) { |
2200 | $volume = volume_rounded($report_totals{Delivered}{Volume}, $report_totals{Delivered}{'Volume-gigs'}); | |
2201 | $total_delivered_count = get_report_total($report_totals{Delivered},'Messages'); | |
2202 | } | |
2203 | else { | |
2204 | $volume = volume_rounded($total_delivered_data, $total_delivered_data_gigs); | |
2205 | } | |
d5692f86 SC |
2206 | |
2207 | my @content=($volume, $total_delivered_count, @delivered_totals); | |
2208 | printf $txt_fh ("$txt_format1\n\n", 'Delivered', @content) if $txt_fh; | |
2209 | printf $htm_fh ("$htm_format1\n\n", 'Delivered', @content) if $htm_fh; | |
2210 | printf $htm_fh "</table>\n" if $htm_fh; | |
2211 | if ($xls_fh) | |
2212 | { | |
2213 | ||
2214 | $row = $row_tablehead+1; | |
2215 | for (my $i=0; $i < scalar(@content); $i++) | |
2216 | { | |
2217 | $ws_global->write($row+$i, 1, $content[$i], $f_default); | |
2218 | } | |
2219 | $row = $row_max; | |
2220 | } | |
059ec3d9 PH |
2221 | } |
2222 | ||
2223 | ||
1b4fe9dd PH |
2224 | ####################################################################### |
2225 | # print_user_patterns() | |
8e669ac1 | 2226 | # |
1b4fe9dd | 2227 | # print_user_patterns(); |
8e669ac1 | 2228 | # |
1b4fe9dd PH |
2229 | # Print the counts of user specified patterns. |
2230 | ####################################################################### | |
059ec3d9 | 2231 | sub print_user_patterns { |
d5692f86 SC |
2232 | my $txt_format1 = " %-18s %6d"; |
2233 | my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%d</td>"; | |
059ec3d9 | 2234 | |
d5692f86 SC |
2235 | if ($txt_fh) { |
2236 | print $txt_fh "User Specified Patterns\n"; | |
2237 | print $txt_fh "-----------------------"; | |
2238 | print $txt_fh "\n Total\n"; | |
059ec3d9 | 2239 | } |
d5692f86 SC |
2240 | if ($htm_fh) { |
2241 | print $htm_fh "<hr><a name=\"patterns\"></a><h2>User Specified Patterns</h2>\n"; | |
2242 | print $htm_fh "<table border=0 width=\"100%\">\n"; | |
2243 | print $htm_fh "<tr><td>\n"; | |
2244 | print $htm_fh "<table border=1>\n"; | |
2245 | print $htm_fh "<tr><th> </th><th>Total</th>\n"; | |
2246 | } | |
2247 | if ($xls_fh) { | |
2248 | $ws_global->write($row++, $col, "User Specified Patterns", $f_header2); | |
2249 | &set_worksheet_line($ws_global, $row++, 1, ["Total"], $f_headertab); | |
059ec3d9 PH |
2250 | } |
2251 | ||
d5692f86 | 2252 | |
059ec3d9 PH |
2253 | my($key); |
2254 | if ($merge_reports) { | |
2255 | # We are getting our data from previous reports. | |
2256 | foreach $key (@user_descriptions) { | |
2257 | my $count = get_report_total($report_totals{patterns}{$key},'Total'); | |
d5692f86 SC |
2258 | printf $txt_fh ("$txt_format1\n",$key,$count) if $txt_fh; |
2259 | printf $htm_fh ("$htm_format1\n",$key,$count) if $htm_fh; | |
2260 | if ($xls_fh) | |
2261 | { | |
2262 | &set_worksheet_line($ws_global, $row++, 0, [$key,$count], $f_default); | |
2263 | } | |
059ec3d9 PH |
2264 | } |
2265 | } | |
2266 | else { | |
2267 | # We are getting our data from mainlog files. | |
2268 | my $user_pattern_index = 0; | |
2269 | foreach $key (@user_descriptions) { | |
d5692f86 SC |
2270 | printf $txt_fh ("$txt_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $txt_fh; |
2271 | printf $htm_fh ("$htm_format1\n",$key,$user_pattern_totals[$user_pattern_index]) if $htm_fh; | |
2272 | if ($xls_fh) | |
2273 | { | |
2274 | &set_worksheet_line($ws_global, $row++, 0, [$key,$user_pattern_totals[$user_pattern_index]]); | |
2275 | } | |
059ec3d9 PH |
2276 | $user_pattern_index++; |
2277 | } | |
2278 | } | |
d5692f86 SC |
2279 | print $txt_fh "\n" if $txt_fh; |
2280 | print $htm_fh "</table>\n\n" if $htm_fh; | |
2281 | if ($xls_fh) | |
2282 | { | |
2283 | ++$row; | |
059ec3d9 | 2284 | } |
059ec3d9 PH |
2285 | } |
2286 | ||
2287 | ||
1b4fe9dd PH |
2288 | ####################################################################### |
2289 | # print_transport(); | |
8e669ac1 | 2290 | # |
1b4fe9dd | 2291 | # print_transport(); |
8e669ac1 | 2292 | # |
1b4fe9dd PH |
2293 | # Print totals by transport. |
2294 | ####################################################################### | |
059ec3d9 | 2295 | sub print_transport { |
059ec3d9 PH |
2296 | my(@chartdatanames); |
2297 | my(@chartdatavals_count); | |
2298 | my(@chartdatavals_vol); | |
d5692f86 SC |
2299 | no integer; #Lose this for charting the data. |
2300 | ||
2301 | my $txt_format1 = " %-18s %6s %6d"; | |
2302 | my $htm_format1 = "<tr><td>%s</td><td align=\"right\">%s</td><td align=\"right\">%d</td>"; | |
059ec3d9 | 2303 | |
d5692f86 SC |
2304 | if ($txt_fh) { |
2305 | print $txt_fh "Deliveries by transport\n"; | |
2306 | print $txt_fh "-----------------------"; | |
2307 | print $txt_fh "\n Volume Messages\n"; | |
059ec3d9 | 2308 | } |
d5692f86 SC |
2309 | if ($htm_fh) { |
2310 | print $htm_fh "<hr><a name=\"transport\"></a><h2>Deliveries by Transport</h2>\n"; | |
2311 | print $htm_fh "<table border=0 width=\"100%\">\n"; | |
2312 | print $htm_fh "<tr><td>\n"; | |
2313 | print $htm_fh "<table border=1>\n"; | |
2314 | print $htm_fh "<tr><th> </th><th>Volume</th><th>Messages</th>\n"; | |
2315 | } | |
2316 | if ($xls_fh) { | |
2317 | $ws_global->write($row++, $col, "Deliveries by transport", $f_header2); | |
2318 | &set_worksheet_line($ws_global, $row++, 1, ["Volume", "Messages"], $f_headertab); | |
059ec3d9 PH |
2319 | } |
2320 | ||
2321 | my($key); | |
2322 | if ($merge_reports) { | |
2323 | # We are getting our data from previous reports. | |
2324 | foreach $key (sort keys %{$report_totals{transport}}) { | |
2325 | my $count = get_report_total($report_totals{transport}{$key},'Messages'); | |
d5692f86 SC |
2326 | my @content=($key, volume_rounded($report_totals{transport}{$key}{Volume}, |
2327 | $report_totals{transport}{$key}{'Volume-gigs'}), $count); | |
059ec3d9 PH |
2328 | push(@chartdatanames, $key); |
2329 | push(@chartdatavals_count, $count); | |
2330 | push(@chartdatavals_vol, $report_totals{transport}{$key}{'Volume-gigs'}*$gig + $report_totals{transport}{$key}{Volume} ); | |
d5692f86 SC |
2331 | printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; |
2332 | printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; | |
2333 | if ($xls_fh) { | |
2334 | &set_worksheet_line($ws_global, $row++, 0, \@content, $f_default); | |
2335 | } | |
059ec3d9 PH |
2336 | } |
2337 | } | |
2338 | else { | |
2339 | # We are getting our data from mainlog files. | |
2340 | foreach $key (sort keys %transported_data) { | |
d5692f86 SC |
2341 | my @content=($key, volume_rounded($transported_data{$key},$transported_data_gigs{$key}), |
2342 | $transported_count{$key}); | |
059ec3d9 PH |
2343 | push(@chartdatanames, $key); |
2344 | push(@chartdatavals_count, $transported_count{$key}); | |
2345 | push(@chartdatavals_vol, $transported_data_gigs{$key}*$gig + $transported_data{$key}); | |
d5692f86 SC |
2346 | printf $txt_fh ("$txt_format1\n", @content) if $txt_fh; |
2347 | printf $htm_fh ("$htm_format1\n", @content) if $htm_fh; | |
2348 | if ($xls_fh) { | |
2349 | &set_worksheet_line($ws_global, $row++, 0, \@content); | |
2350 | } | |
059ec3d9 PH |
2351 | } |
2352 | } | |
d5692f86 SC |
2353 | print $txt_fh "\n" if $txt_fh; |
2354 | if ($htm_fh) { | |
2355 | print $htm_fh "</table>\n"; | |
2356 | print $htm_fh "</td><td>\n"; | |
059ec3d9 PH |
2357 | if ($HAVE_GD_Graph_pie && $charts) |
2358 | { | |
2359 | # calculate the graph | |
2360 | my @data = ( | |
2361 | \@chartdatanames, | |
2362 | \@chartdatavals_count | |
2363 | ); | |
2364 | my $graph = GD::Graph::pie->new(200, 200); | |
2365 | $graph->set( | |
2366 | x_label => 'Transport', | |
2367 | y_label => 'Messages', | |
2368 | title => 'By count', | |
2369 | ); | |
2370 | my $gd = $graph->plot(\@data) or warn($graph->error); | |
2371 | if ($gd) { | |
d5692f86 SC |
2372 | open(IMG, ">$chartdir/transports_count.png") or die "Could not write $chartdir/transports_count.png: $!\n"; |
2373 | binmode IMG; | |
2374 | print IMG $gd->png; | |
2375 | close IMG; | |
2376 | print $htm_fh "<img src=\"$chartrel/transports_count.png\">"; | |
059ec3d9 PH |
2377 | } |
2378 | } | |
d5692f86 | 2379 | print $htm_fh "</td><td>\n"; |
059ec3d9 PH |
2380 | |
2381 | if ($HAVE_GD_Graph_pie && $charts) { | |
2382 | my @data = ( | |
2383 | \@chartdatanames, | |
2384 | \@chartdatavals_vol | |
2385 | ); | |
2386 | my $graph = GD::Graph::pie->new(200, 200); | |
2387 | $graph->set( | |
2388 | title => 'By volume', | |
2389 | ); | |
2390 | my $gd = $graph->plot(\@data) or warn($graph->error); | |
2391 | if ($gd) { | |
d5692f86 SC |
2392 | open(IMG, ">$chartdir/transports_vol.png") or die "Could not write $chartdir/transports_count.png: $!\n"; |
2393 | binmode IMG; | |
2394 | print IMG $gd->png; | |
2395 | close IMG; | |
2396 | print $htm_fh "<img src=\"$chartrel/transports_vol.png\">"; | |
059ec3d9 PH |
2397 | } |
2398 | } | |
d5692f86 SC |
2399 | print $htm_fh "</td></tr></table>\n\n"; |
2400 | } | |
2401 | if ($xls_fh) { | |
2402 | $row++; | |
059ec3d9 | 2403 | } |
d5692f86 | 2404 | |
059ec3d9 PH |
2405 | } |
2406 | ||
2407 | ||
2408 | ||
1b4fe9dd PH |
2409 | ####################################################################### |
2410 | # print_relay(); | |
8e669ac1 | 2411 | # |
1b4fe9dd | 2412 | # print_relay(); |
8e669ac1 | 2413 | # |
1b4fe9dd PH |
2414 | # Print our totals by relay. |
2415 | ####################################################################### | |
059ec3d9 | 2416 | sub print_relay { |
d5692f86 | 2417 | my $row_print_relay=1; |
059ec3d9 | 2418 | my $temp = "Relayed messages"; |
d5692f86 | 2419 | print $htm_fh "<hr><a name=\"$temp\"></a><h2>$temp</h2>\n" if $htm_fh; |
059ec3d9 PH |
2420 | if (scalar(keys %relayed) > 0 || $relayed_unshown > 0) { |
2421 | my $shown = 0; | |
2422 | my $spacing = ""; | |
d5692f86 SC |
2423 | my $txt_format = "%7d %s\n => %s\n"; |
2424 | my $htm_format = "<tr><td align=\"right\">%d</td><td>%s</td><td>%s</td>\n"; | |
059ec3d9 | 2425 | |
d5692f86 SC |
2426 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; |
2427 | if ($htm_fh) { | |
2428 | print $htm_fh "<table border=1>\n"; | |
2429 | print $htm_fh "<tr><th>Count</th><th>From</th><th>To</th>\n"; | |
059ec3d9 | 2430 | } |
d5692f86 SC |
2431 | if ($xls_fh) { |
2432 | $ws_relayed->write($row_print_relay++, $col, $temp, $f_header2); | |
2433 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, ["Count", "From", "To"], $f_headertab); | |
059ec3d9 PH |
2434 | } |
2435 | ||
d5692f86 | 2436 | |
059ec3d9 PH |
2437 | my($key); |
2438 | foreach $key (sort keys %relayed) { | |
2439 | my $count = $relayed{$key}; | |
2440 | $shown += $count; | |
2441 | $key =~ s/[HA]=//g; | |
2442 | my($one,$two) = split(/=> /, $key); | |
d5692f86 SC |
2443 | my @content=($count, $one, $two); |
2444 | printf $txt_fh ($txt_format, @content) if $txt_fh; | |
2445 | printf $htm_fh ($htm_format, @content) if $htm_fh; | |
2446 | if ($xls_fh) | |
2447 | { | |
2448 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, \@content); | |
2449 | } | |
059ec3d9 PH |
2450 | $spacing = "\n"; |
2451 | } | |
d5692f86 SC |
2452 | |
2453 | print $htm_fh "</table>\n<p>\n" if $htm_fh; | |
2454 | print $txt_fh "${spacing}Total: $shown (plus $relayed_unshown unshown)\n\n" if $txt_fh; | |
2455 | print $htm_fh "${spacing}Total: $shown (plus $relayed_unshown unshown)\n\n" if $htm_fh; | |
2456 | if ($xls_fh) | |
2457 | { | |
2458 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, [$shown, "Sum of shown" ]); | |
2459 | &set_worksheet_line($ws_relayed, $row_print_relay++, 0, [$relayed_unshown, "unshown"]); | |
2460 | $row_print_relay++; | |
2461 | } | |
059ec3d9 PH |
2462 | } |
2463 | else { | |
d5692f86 SC |
2464 | print $txt_fh "No relayed messages\n-------------------\n\n" if $txt_fh; |
2465 | print $htm_fh "No relayed messages\n\n" if $htm_fh; | |
2466 | if ($xls_fh) | |
2467 | { | |
2468 | $row_print_relay++; | |
2469 | } | |
059ec3d9 | 2470 | } |
059ec3d9 PH |
2471 | } |
2472 | ||
2473 | ||
2474 | ||
1b4fe9dd PH |
2475 | ####################################################################### |
2476 | # print_errors(); | |
8e669ac1 | 2477 | # |
1b4fe9dd | 2478 | # print_errors(); |
8e669ac1 | 2479 | # |
1b4fe9dd PH |
2480 | # Print our errors. In HTML, we display them as a list rather than a table - |
2481 | # Netscape doesn't like large tables! | |
2482 | ####################################################################### | |
059ec3d9 PH |
2483 | sub print_errors { |
2484 | my $total_errors = 0; | |
d5692f86 | 2485 | $row=1; |
059ec3d9 PH |
2486 | |
2487 | if (scalar(keys %errors_count) != 0) { | |
2488 | my $temp = "List of errors"; | |
d5692f86 SC |
2489 | my $htm_format = "<li>%d - %s\n"; |
2490 | ||
2491 | printf $txt_fh ("%s\n%s\n\n", $temp, "-" x length($temp)) if $txt_fh; | |
2492 | if ($htm_fh) { | |
2493 | print $htm_fh "<hr><a name=\"errors\"></a><h2>$temp</h2>\n"; | |
2494 | print $htm_fh "<ul><li><b>Count - Error</b>\n"; | |
059ec3d9 | 2495 | } |
d5692f86 SC |
2496 | if ($xls_fh) |
2497 | { | |
2498 | $ws_errors->write($row++, 0, $temp, $f_header2); | |
2499 | &set_worksheet_line($ws_errors, $row++, 0, ["Count", "Error"], $f_headertab); | |
059ec3d9 PH |
2500 | } |
2501 | ||
d5692f86 | 2502 | |
059ec3d9 PH |
2503 | my($key); |
2504 | foreach $key (sort keys %errors_count) { | |
2505 | my $text = $key; | |
2506 | chomp($text); | |
d5692f86 | 2507 | $text =~ s/\s\s+/ /g; #Convert multiple spaces to a single space. |
059ec3d9 | 2508 | $total_errors += $errors_count{$key}; |
d5692f86 SC |
2509 | |
2510 | if ($txt_fh) { | |
2511 | printf $txt_fh ("%5d ", $errors_count{$key}); | |
2512 | my $text_remaining = $text; | |
2513 | while (length($text_remaining) > 65) { | |
2514 | my($first,$rest) = $text_remaining =~ /(.{50}\S*)\s+(.+)/; | |
2515 | last if !$first; | |
2516 | printf $txt_fh ("%s\n\t ", $first); | |
2517 | $text_remaining = $rest; | |
2518 | } | |
2519 | printf $txt_fh ("%s\n\n", $text_remaining); | |
2520 | } | |
2521 | ||
2522 | if ($htm_fh) { | |
8e669ac1 | 2523 | |
059ec3d9 PH |
2524 | #Translate HTML tag characters. Sergey Sholokh. |
2525 | $text =~ s/\</\<\;/g; | |
2526 | $text =~ s/\>/\>\;/g; | |
2527 | ||
d5692f86 | 2528 | printf $htm_fh ($htm_format,$errors_count{$key},$text); |
059ec3d9 | 2529 | } |
d5692f86 SC |
2530 | if ($xls_fh) |
2531 | { | |
2532 | &set_worksheet_line($ws_errors, $row++, 0, [$errors_count{$key},$text]); | |
059ec3d9 PH |
2533 | } |
2534 | } | |
059ec3d9 PH |
2535 | |
2536 | $temp = "Errors encountered: $total_errors"; | |
d5692f86 SC |
2537 | |
2538 | if ($txt_fh) { | |
2539 | print $txt_fh $temp, "\n"; | |
2540 | print $txt_fh "-" x length($temp),"\n"; | |
2541 | } | |
2542 | if ($htm_fh) { | |
2543 | print $htm_fh "</ul>\n<p>\n"; | |
2544 | print $htm_fh $temp, "\n"; | |
2545 | } | |
2546 | if ($xls_fh) | |
2547 | { | |
2548 | &set_worksheet_line($ws_errors, $row++, 0, [$total_errors, "Sum of Errors encountered"]); | |
2549 | } | |
059ec3d9 PH |
2550 | } |
2551 | ||
2552 | } | |
2553 | ||
2554 | ||
1b4fe9dd PH |
2555 | ####################################################################### |
2556 | # parse_old_eximstat_reports(); | |
8e669ac1 | 2557 | # |
1b4fe9dd | 2558 | # parse_old_eximstat_reports($fh); |
8e669ac1 | 2559 | # |
1b4fe9dd | 2560 | # Parse old eximstat output so we can merge daily stats to weekly stats and weekly to monthly etc. |
8e669ac1 | 2561 | # |
1b4fe9dd PH |
2562 | # To test that the merging still works after changes, do something like the following. |
2563 | # All the diffs should produce no output. | |
8e669ac1 | 2564 | # |
1b4fe9dd PH |
2565 | # options='-bydomain -byemail -byhost -byedomain' |
2566 | # options="$options -pattern 'Completed Messages' /Completed/" | |
2567 | # options="$options -pattern 'Received Messages' /<=/" | |
8e669ac1 | 2568 | # |
1b4fe9dd PH |
2569 | # ./eximstats $options mainlog > mainlog.txt |
2570 | # ./eximstats $options -merge mainlog.txt > mainlog.2.txt | |
2571 | # diff mainlog.txt mainlog.2.txt | |
8e669ac1 | 2572 | # |
1b4fe9dd PH |
2573 | # ./eximstats $options -html mainlog > mainlog.html |
2574 | # ./eximstats $options -merge -html mainlog.txt > mainlog.2.html | |
2575 | # diff mainlog.html mainlog.2.html | |
8e669ac1 | 2576 | # |
1b4fe9dd PH |
2577 | # ./eximstats $options -merge mainlog.html > mainlog.3.txt |
2578 | # diff mainlog.txt mainlog.3.txt | |
8e669ac1 | 2579 | # |
1b4fe9dd PH |
2580 | # ./eximstats $options -merge -html mainlog.html > mainlog.3.html |
2581 | # diff mainlog.html mainlog.3.html | |
8e669ac1 | 2582 | # |
1b4fe9dd PH |
2583 | # ./eximstats $options -nvr mainlog > mainlog.nvr.txt |
2584 | # ./eximstats $options -merge mainlog.nvr.txt > mainlog.4.txt | |
2585 | # diff mainlog.txt mainlog.4.txt | |
8e669ac1 | 2586 | # |
1b4fe9dd PH |
2587 | # # double_mainlog.txt should have twice the values that mainlog.txt has. |
2588 | # ./eximstats $options mainlog mainlog > double_mainlog.txt | |
2589 | ####################################################################### | |
059ec3d9 PH |
2590 | sub parse_old_eximstat_reports { |
2591 | my($fh) = @_; | |
2592 | ||
2593 | my(%league_table_value_entered, %league_table_value_was_zero, %table_order); | |
2594 | ||
2595 | while (<$fh>) { | |
d5692f86 | 2596 | PARSE_OLD_REPORT_LINE: |
059ec3d9 PH |
2597 | if (/Exim statistics from ([\d\-]+ [\d:]+(\s+[\+\-]\d+)?) to ([\d\-]+ [\d:]+(\s+[\+\-]\d+)?)/) { |
2598 | $begin = $1 if ($1 lt $begin); | |
2599 | $end = $3 if ($3 gt $end); | |
2600 | } | |
2601 | elsif (/Grand total summary/) { | |
2602 | # Fill in $report_totals{Received|Delivered}{Volume|Messages|Hosts|Domains|...|Delayed|DelayedPercent|Failed|FailedPercent} | |
2603 | my(@fields); | |
2604 | while (<$fh>) { | |
d5692f86 SC |
2605 | $_ = html2txt($_); #Convert general HTML markup to text. |
2606 | s/At least one addr//g; #Another part of the HTML output we don't want. | |
059ec3d9 PH |
2607 | |
2608 | # TOTAL Volume Messages Hosts Domains Delayed Failed | |
2609 | # Received 26MB 237 177 23 8 3.4% 28 11.8% | |
2610 | # Delivered 13MB 233 99 88 | |
d5692f86 SC |
2611 | if (/TOTAL\s+(.*?)\s*$/) { |
2612 | @fields = split(/\s+/,$1); | |
059ec3d9 | 2613 | #Delayed and Failed have two columns each, so add the extra field names in. |
d5692f86 SC |
2614 | splice(@fields,-1,1,'DelayedPercent','Failed','FailedPercent'); |
2615 | } | |
2616 | elsif (/(Received|Delivered)\s+(.*?)\s*$/) { | |
2617 | print STDERR "Parsing $_" if $debug; | |
2618 | add_to_totals($report_totals{$1},\@fields,$2); | |
2619 | } | |
2620 | last if (/Delivered/); #Last line of this section. | |
059ec3d9 PH |
2621 | } |
2622 | } | |
2623 | ||
2624 | elsif (/User Specified Patterns/i) { | |
2625 | #User Specified Patterns | |
2626 | #----------------------- | |
2627 | # Total | |
2628 | # Description 85 | |
2629 | ||
d5692f86 | 2630 | while (<$fh>) { last if (/Total/); } #Wait until we get the table headers. |
059ec3d9 | 2631 | while (<$fh>) { |
d5692f86 SC |
2632 | print STDERR "Parsing $_" if $debug; |
2633 | $_ = html2txt($_); #Convert general HTML markup to text. | |
2634 | if (/^\s*(.*?)\s+(\d+)\s*$/) { | |
2635 | $report_totals{patterns}{$1} = {} unless (defined $report_totals{patterns}{$1}); | |
2636 | add_to_totals($report_totals{patterns}{$1},['Total'],$2); | |
2637 | } | |
2638 | last if (/^\s*$/); #Finished if we have a blank line. | |
059ec3d9 PH |
2639 | } |
2640 | } | |
2641 | ||
2642 | elsif (/Deliveries by transport/i) { | |
2643 | #Deliveries by transport | |
2644 | #----------------------- | |
2645 | # Volume Messages | |
2646 | # :blackhole: 70KB 51 | |
2647 | # address_pipe 655KB 1 | |
2648 | # smtp 11MB 151 | |
2649 | ||
d5692f86 | 2650 | while (<$fh>) { last if (/Volume/); } #Wait until we get the table headers. |
059ec3d9 | 2651 | while (<$fh>) { |
d5692f86 SC |
2652 | print STDERR "Parsing $_" if $debug; |
2653 | $_ = html2txt($_); #Convert general HTML markup to text. | |
2654 | if (/(\S+)\s+(\d+\S*\s+\d+)/) { | |
2655 | $report_totals{transport}{$1} = {} unless (defined $report_totals{transport}{$1}); | |
2656 | add_to_totals($report_totals{transport}{$1},['Volume','Messages'],$2); | |
2657 | } | |
2658 | last if (/^\s*$/); #Finished if we have a blank line. | |
059ec3d9 PH |
2659 | } |
2660 | } | |
2661 | elsif (/(Messages received|Deliveries) per/) { | |
2662 | # Messages received per hour (each dot is 2 messages) | |
2663 | #--------------------------------------------------- | |
2664 | # | |
2665 | #00-01 106 ..................................................... | |
2666 | #01-02 103 ................................................... | |
2667 | ||
2668 | # Set a pointer to the interval array so we can use the same code | |
2669 | # block for both messages received and delivered. | |
2670 | my $interval_aref = ($1 eq 'Deliveries') ? \@delivered_interval_count : \@received_interval_count; | |
2671 | my $reached_table = 0; | |
2672 | while (<$fh>) { | |
d5692f86 SC |
2673 | $reached_table = 1 if (/^00/); |
2674 | next unless $reached_table; | |
2675 | print STDERR "Parsing $_" if $debug; | |
2676 | if (/^(\d+):(\d+)\s+(\d+)/) { #hh:mm start time format ? | |
2677 | $$interval_aref[($1*60 + $2)/$hist_interval] += $3 if $hist_opt; | |
2678 | } | |
2679 | elsif (/^(\d+)-(\d+)\s+(\d+)/) { #hh-hh start-end time format ? | |
2680 | $$interval_aref[($1*60)/$hist_interval] += $3 if $hist_opt; | |
2681 | } | |
2682 | else { #Finished the table ? | |
2683 | last; | |
2684 | } | |
059ec3d9 PH |
2685 | } |
2686 | } | |
2687 | ||
2688 | elsif (/Time spent on the queue: (all messages|messages with at least one remote delivery)/) { | |
2689 | #Time spent on the queue: all messages | |
2690 | #------------------------------------- | |
2691 | # | |
2692 | #Under 1m 217 91.9% 91.9% | |
2693 | # 5m 2 0.8% 92.8% | |
2694 | # 3h 8 3.4% 96.2% | |
2695 | # 6h 7 3.0% 99.2% | |
2696 | # 12h 2 0.8% 100.0% | |
2697 | ||
2698 | # Set a pointer to the queue bin so we can use the same code | |
2699 | # block for both all messages and remote deliveries. | |
2700 | my $bin_aref = ($1 eq 'all messages') ? \@queue_bin : \@remote_queue_bin; | |
2701 | my $reached_table = 0; | |
2702 | while (<$fh>) { | |
d5692f86 SC |
2703 | $_ = html2txt($_); #Convert general HTML markup to text. |
2704 | $reached_table = 1 if (/^\s*Under/); | |
2705 | next unless $reached_table; | |
2706 | my $previous_seconds_on_queue = 0; | |
2707 | if (/^\s*(Under|Over|)\s+(\d+[smhdw])\s+(\d+)/) { | |
2708 | print STDERR "Parsing $_" if $debug; | |
2709 | my($modifier,$formated_time,$count) = ($1,$2,$3); | |
2710 | my $seconds = unformat_time($formated_time); | |
2711 | my $time_on_queue = ($seconds + $previous_seconds_on_queue) / 2; | |
2712 | $previous_seconds_on_queue = $seconds; | |
2713 | $time_on_queue = $seconds * 2 if ($modifier eq 'Over'); | |
2714 | my($i); | |
2715 | for ($i = 0; $i <= $#queue_times; $i++) { | |
2716 | if ($time_on_queue < $queue_times[$i]) { | |
2717 | $$bin_aref[$i] += $count; | |
2718 | last; | |
2719 | } | |
2720 | } | |
059ec3d9 PH |
2721 | # There's only one counter for messages going over the queue |
2722 | # times so make sure we only count it once. | |
2723 | $queue_more_than += $count if (($bin_aref == \@queue_bin) && ($i > $#queue_times)); | |
d5692f86 SC |
2724 | } |
2725 | else { | |
2726 | last; #Finished the table ? | |
2727 | } | |
059ec3d9 PH |
2728 | } |
2729 | } | |
2730 | ||
2731 | elsif (/Relayed messages/) { | |
2732 | #Relayed messages | |
2733 | #---------------- | |
2734 | # | |
2735 | # 1 addr.domain.com [1.2.3.4] a.user@domain.com | |
2736 | # => addr2.domain2.com [5.6.7.8] a2.user2@domain2.com | |
2737 | # | |
2738 | #<tr><td align="right">1</td><td>addr.domain.com [1.2.3.4] a.user@domain.com </td><td>addr2.domain2.com [5.6.7.8] a2.user2@domain2.com</td> | |
2739 | ||
2740 | my $reached_table = 0; | |
2741 | my($count,$sender); | |
2742 | while (<$fh>) { | |
d5692f86 SC |
2743 | unless ($reached_table) { |
2744 | last if (/No relayed messages/); | |
2745 | $reached_table = 1 if (/^\s*\d/ || />\d+</); | |
2746 | next unless $reached_table; | |
2747 | } | |
2748 | if (/>(\d+)<.td><td>(.*?) ?<.td><td>(.*?)</) { | |
2749 | update_relayed($1,$2,$3); | |
2750 | } | |
2751 | elsif (/^\s*(\d+)\s+(.*?)\s*$/) { | |
2752 | ($count,$sender) = ($1,$2); | |
2753 | } | |
2754 | elsif (/=>\s+(.*?)\s*$/) { | |
2755 | update_relayed($count,$sender,$1); | |
2756 | } | |
2757 | else { | |
2758 | last; #Finished the table ? | |
2759 | } | |
059ec3d9 PH |
2760 | } |
2761 | } | |
2762 | ||
2763 | elsif (/Top (.*?) by (message count|volume)/) { | |
2764 | #Top 50 sending hosts by message count | |
2765 | #------------------------------------- | |
2766 | # | |
2767 | # 48 1468KB local | |
50adf73a SC |
2768 | # Could also have average values for HTML output. |
2769 | # 48 1468KB 30KB local | |
2770 | ||
059ec3d9 PH |
2771 | my($category,$by_count_or_volume) = ($1,$2); |
2772 | ||
2773 | #As we show 2 views of each table (by count and by volume), | |
2774 | #most (but not all) entries will appear in both tables. | |
2775 | #Set up a hash to record which entries we have already seen | |
2776 | #and one to record which ones we are seeing for the first time. | |
2777 | if ($by_count_or_volume =~ /count/) { | |
d5692f86 SC |
2778 | undef %league_table_value_entered; |
2779 | undef %league_table_value_was_zero; | |
2780 | undef %table_order; | |
059ec3d9 PH |
2781 | } |
2782 | ||
2783 | #As this section processes multiple different table categories, | |
2784 | #set up pointers to the hashes to be updated. | |
2785 | my($count_href,$data_href,$data_gigs_href); | |
2786 | if ($category =~ /local sender/) { | |
2787 | $count_href = \%received_count_user; | |
d5692f86 SC |
2788 | $data_href = \%received_data_user; |
2789 | $data_gigs_href = \%received_data_gigs_user; | |
059ec3d9 PH |
2790 | } |
2791 | elsif ($category =~ /sending (\S+?)s?\b/) { | |
2792 | #Top 50 sending (host|domain|email|edomain)s | |
2793 | #Top sending (host|domain|email|edomain) | |
2794 | $count_href = \%{$received_count{"\u$1"}}; | |
d5692f86 SC |
2795 | $data_href = \%{$received_data{"\u$1"}}; |
2796 | $data_gigs_href = \%{$received_data_gigs{"\u$1"}}; | |
059ec3d9 PH |
2797 | } |
2798 | elsif ($category =~ /local destination/) { | |
2799 | $count_href = \%delivered_count_user; | |
d5692f86 SC |
2800 | $data_href = \%delivered_data_user; |
2801 | $data_gigs_href = \%delivered_data_gigs_user; | |
059ec3d9 PH |
2802 | } |
2803 | elsif ($category =~ /(\S+) destination/) { | |
2804 | #Top 50 (host|domain|email|edomain) destinations | |
2805 | #Top (host|domain|email|edomain) destination | |
2806 | $count_href = \%{$delivered_count{"\u$1"}}; | |
d5692f86 SC |
2807 | $data_href = \%{$delivered_data{"\u$1"}}; |
2808 | $data_gigs_href = \%{$delivered_data_gigs{"\u$1"}}; | |
059ec3d9 PH |
2809 | } |
2810 | ||
2811 | my $reached_table = 0; | |
2812 | while (<$fh>) { | |
d5692f86 SC |
2813 | # Watch out for empty tables. |
2814 | goto PARSE_OLD_REPORT_LINE if (/<h2>/ or /^[a-zA-Z]/); | |
2815 | ||
2816 | $_ = html2txt($_); #Convert general HTML markup to text. | |
2817 | ||
2818 | ||
2819 | $reached_table = 1 if (/^\s*\d/); | |
2820 | next unless $reached_table; | |
50adf73a SC |
2821 | |
2822 | # Remove optional 'average value' column. | |
d5692f86 | 2823 | s/^\s*(\d+)\s+(\S+)\s+(\d+(KB|MB|GB|\b)\s+)/$1 $2 /; |
50adf73a | 2824 | |
d5692f86 SC |
2825 | if (/^\s*(\d+)\s+(\S+)\s*(.*?)\s*$/) { |
2826 | my($count,$rounded_volume,$entry) = ($1,$2,$3); | |
059ec3d9 PH |
2827 | #Note: $entry fields can be both null and can contain spaces. |
2828 | ||
d5692f86 SC |
2829 | #Add the entry into the %table_order hash if it has a rounded volume (KB/MB/GB). |
2830 | push(@{$table_order{$rounded_volume}{$by_count_or_volume}},$entry) if ($rounded_volume =~ /\D/); | |
059ec3d9 PH |
2831 | |
2832 | unless ($league_table_value_entered{$entry}) { | |
d5692f86 SC |
2833 | $league_table_value_entered{$entry} = 1; |
2834 | unless ($$count_href{$entry}) { | |
2835 | $$count_href{$entry} = 0; | |
2836 | $$data_href{$entry} = 0; | |
2837 | $$data_gigs_href{$entry} = 0; | |
2838 | $league_table_value_was_zero{$entry} = 1; | |
2839 | } | |
2840 | ||
2841 | $$count_href{$entry} += $count; | |
059ec3d9 | 2842 | #Add the rounded value to the data and data_gigs hashes. |
d5692f86 SC |
2843 | un_round($rounded_volume,\$$data_href{$entry},\$$data_gigs_href{$entry}); |
2844 | print STDERR "$category by $by_count_or_volume: added $count,$rounded_volume to $entry\n" if $debug; | |
2845 | } | |
2846 | } | |
2847 | else { #Finished the table ? | |
2848 | if ($by_count_or_volume =~ /volume/) { | |
2849 | #Add a few bytes to appropriate entries to preserve the order. | |
2850 | ||
2851 | my($rounded_volume); | |
2852 | foreach $rounded_volume (keys %table_order) { | |
2853 | #For each rounded volume, we want to create a list which has things | |
2854 | #ordered from the volume table at the front, and additional things | |
2855 | #from the count table ordered at the back. | |
2856 | @{$table_order{$rounded_volume}{volume}} = () unless defined $table_order{$rounded_volume}{volume}; | |
2857 | @{$table_order{$rounded_volume}{'message count'}} = () unless defined $table_order{$rounded_volume}{'message count'}; | |
2858 | my(@order,%mark); | |
2859 | map {$mark{$_} = 1} @{$table_order{$rounded_volume}{volume}}; | |
2860 | @order = @{$table_order{$rounded_volume}{volume}}; | |
2861 | map {push(@order,$_)} grep(!$mark{$_},@{$table_order{$rounded_volume}{'message count'}}); | |
2862 | ||
2863 | my $bonus_bytes = $#order; | |
2864 | $bonus_bytes = 511 if ($bonus_bytes > 511); #Don't go over the half-K boundary! | |
2865 | while (@order and ($bonus_bytes > 0)) { | |
2866 | my $entry = shift(@order); | |
2867 | if ($league_table_value_was_zero{$entry}) { | |
2868 | $$data_href{$entry} += $bonus_bytes; | |
2869 | print STDERR "$category by $by_count_or_volume: added $bonus_bytes bonus bytes to $entry\n" if $debug; | |
2870 | } | |
2871 | $bonus_bytes--; | |
2872 | } | |
2873 | } | |
2874 | } | |
2875 | ||
2876 | last; | |
2877 | } | |
059ec3d9 PH |
2878 | } |
2879 | } | |
2880 | elsif (/List of errors/) { | |
2881 | #List of errors | |
2882 | #-------------- | |
2883 | # | |
2884 | # 1 07904931641@one2one.net R=external T=smtp: SMTP error | |
2885 | # from remote mailer after RCPT TO:<07904931641@one2one.net>: | |
2886 | # host mail.one2one.net [193.133.192.24]: 550 User unknown | |
2887 | # | |
2888 | #<li>1 - ally.dufc@dunbar.org.uk R=external T=smtp: SMTP error from remote mailer after RCPT TO:<ally.dufc@dunbar.org.uk>: host mail.dunbar.org.uk [216.167.89.88]: 550 Unknown local part ally.dufc in <ally.dufc@dunbar.org.uk> | |
2889 | ||
2890 | ||
2891 | my $reached_table = 0; | |
2892 | my($count,$error,$blanks); | |
2893 | while (<$fh>) { | |
d5692f86 SC |
2894 | $reached_table = 1 if (/^( *|<li>)(\d+)/); |
2895 | next unless $reached_table; | |
059ec3d9 | 2896 | |
d5692f86 SC |
2897 | s/^<li>(\d+) -/$1/; #Convert an HTML line to a text line. |
2898 | $_ = html2txt($_); #Convert general HTML markup to text. | |
059ec3d9 | 2899 | |
d5692f86 SC |
2900 | if (/\t\s*(.*)/) { |
2901 | $error .= ' ' . $1; #Join a multiline error. | |
2902 | } | |
2903 | elsif (/^\s*(\d+)\s+(.*)/) { | |
2904 | if ($error) { | |
059ec3d9 | 2905 | #Finished with a previous multiline error so save it. |
d5692f86 SC |
2906 | $errors_count{$error} = 0 unless $errors_count{$error}; |
2907 | $errors_count{$error} += $count; | |
2908 | } | |
2909 | ($count,$error) = ($1,$2); | |
2910 | } | |
2911 | elsif (/Errors encountered/) { | |
2912 | if ($error) { | |
059ec3d9 | 2913 | #Finished the section, so save our stored last error. |
d5692f86 SC |
2914 | $errors_count{$error} = 0 unless $errors_count{$error}; |
2915 | $errors_count{$error} += $count; | |
2916 | } | |
2917 | last; | |
2918 | } | |
059ec3d9 PH |
2919 | } |
2920 | } | |
2921 | ||
2922 | } | |
2923 | } | |
2924 | ||
2925 | ||
2926 | ||
1b4fe9dd PH |
2927 | ####################################################################### |
2928 | # update_relayed(); | |
8e669ac1 | 2929 | # |
1b4fe9dd | 2930 | # update_relayed($count,$sender,$recipient); |
8e669ac1 | 2931 | # |
1b4fe9dd PH |
2932 | # Adds an entry into the %relayed hash. Currently only used when |
2933 | # merging reports. | |
2934 | ####################################################################### | |
059ec3d9 PH |
2935 | sub update_relayed { |
2936 | my($count,$sender,$recipient) = @_; | |
2937 | ||
2938 | #When generating the key, put in the 'H=' and 'A=' which can be used | |
2939 | #in searches. | |
2940 | my $key = "H=$sender => H=$recipient"; | |
2941 | $key =~ s/ ([^=\s]+\@\S+|<>)/ A=$1/g; | |
2942 | if (!defined $relay_pattern || $key !~ /$relay_pattern/o) { | |
2943 | $relayed{$key} = 0 if !defined $relayed{$key}; | |
2944 | $relayed{$key} += $count; | |
2945 | } | |
2946 | else { | |
2947 | $relayed_unshown += $count; | |
2948 | } | |
2949 | } | |
2950 | ||
2951 | ||
1b4fe9dd PH |
2952 | ####################################################################### |
2953 | # add_to_totals(); | |
8e669ac1 | 2954 | # |
1b4fe9dd | 2955 | # add_to_totals(\%totals,\@keys,$values); |
8e669ac1 | 2956 | # |
1b4fe9dd PH |
2957 | # Given a line of space seperated values, add them into the provided hash using @keys |
2958 | # as the hash keys. | |
8e669ac1 | 2959 | # |
1b4fe9dd PH |
2960 | # If the value contains a '%', then the value is set rather than added. Otherwise, we |
2961 | # convert the value to bytes and gigs. The gigs get added to I<Key>-gigs. | |
2962 | ####################################################################### | |
059ec3d9 PH |
2963 | sub add_to_totals { |
2964 | my($totals_href,$keys_aref,$values) = @_; | |
2965 | my(@values) = split(/\s+/,$values); | |
d5692f86 | 2966 | my(@keys) = @$keys_aref; #Make a copy as we destroy the one we use. |
059ec3d9 PH |
2967 | my($value); |
2968 | foreach $value (@values) { | |
2969 | my $key = shift(@keys) or next; | |
2970 | if ($value =~ /%/) { | |
2971 | $$totals_href{$key} = $value; | |
2972 | } | |
2973 | else { | |
2974 | $$totals_href{$key} = 0 unless ($$totals_href{$key}); | |
2975 | $$totals_href{"$key-gigs"} = 0 unless ($$totals_href{"$key-gigs"}); | |
2976 | un_round($value, \$$totals_href{$key}, \$$totals_href{"$key-gigs"}); | |
2977 | print STDERR "Added $value to $key - $$totals_href{$key} , " . $$totals_href{"$key-gigs"} . "GB.\n" if $debug; | |
2978 | } | |
2979 | } | |
2980 | } | |
2981 | ||
1b4fe9dd PH |
2982 | ####################################################################### |
2983 | # get_report_total(); | |
8e669ac1 | 2984 | # |
1b4fe9dd | 2985 | # $total = get_report_total(\%hash,$key); |
8e669ac1 | 2986 | # |
1b4fe9dd | 2987 | # If %hash contains values split into Units and Gigs, we calculate and return |
8e669ac1 | 2988 | # |
1b4fe9dd PH |
2989 | # $hash{$key} + 1024*1024*1024 * $hash{"${key}-gigs"} |
2990 | ####################################################################### | |
059ec3d9 PH |
2991 | sub get_report_total { |
2992 | no integer; | |
2993 | my($hash_ref,$key) = @_; | |
2994 | if ($$hash_ref{"${key}-gigs"}) { | |
2995 | return $$hash_ref{$key} + $gig * $$hash_ref{"${key}-gigs"}; | |
2996 | } | |
2997 | return $$hash_ref{$key} || 0; | |
2998 | } | |
2999 | ||
1b4fe9dd PH |
3000 | ####################################################################### |
3001 | # html2txt(); | |
8e669ac1 | 3002 | # |
1b4fe9dd | 3003 | # $text_line = html2txt($html_line); |
8e669ac1 | 3004 | # |
1b4fe9dd PH |
3005 | # Convert a line from html to text. Currently we just convert HTML tags to spaces |
3006 | # and convert >, <, and tags back. | |
3007 | ####################################################################### | |
059ec3d9 PH |
3008 | sub html2txt { |
3009 | ($_) = @_; | |
3010 | ||
3011 | # Convert HTML tags to spacing. Note that the reports may contain <Userid> and | |
3012 | # <Userid@Domain> words, so explicitly specify the HTML tags we will remove | |
3013 | # (the ones used by this program). If someone is careless enough to have their | |
3014 | # Userid the same as an HTML tag, there's not much we can do about it. | |
3015 | s/<\/?(html|head|title|body|h\d|ul|li|a\s+|table|tr|td|th|pre|hr|p|br)\b.*?>/ /og; | |
3016 | ||
3017 | s/\<\;/\</og; #Convert '<' to '<'. | |
3018 | s/\>\;/\>/og; #Convert '>' to '>'. | |
3019 | s/\ \;/ /og; #Convert ' ' to ' '. | |
3020 | return($_); | |
3021 | } | |
3022 | ||
1b4fe9dd PH |
3023 | ####################################################################### |
3024 | # get_next_arg(); | |
8e669ac1 | 3025 | # |
1b4fe9dd | 3026 | # $arg = get_next_arg(); |
8e669ac1 | 3027 | # |
1b4fe9dd PH |
3028 | # Because eximstats arguments are often passed as variables, |
3029 | # we can't rely on shell parsing to deal with quotes. This | |
3030 | # subroutine returns $ARGV[1] and does a shift. If $ARGV[1] | |
3031 | # starts with a quote (' or "), and doesn't end in one, then | |
3032 | # we append the next argument to it and shift again. We repeat | |
3033 | # until we've got all of the argument. | |
8e669ac1 | 3034 | # |
1b4fe9dd PH |
3035 | # This isn't perfect as all white space gets reduced to one space, |
3036 | # but it's as good as we can get! If it's esential that spacing | |
3037 | # be preserved precisely, then you get that by not using shell | |
3038 | # variables. | |
3039 | ####################################################################### | |
059ec3d9 PH |
3040 | sub get_next_arg { |
3041 | my $arg = ''; | |
3042 | my $matched_pattern = 0; | |
3043 | while ($ARGV[1]) { | |
3044 | $arg .= ' ' if $arg; | |
3045 | $arg .= $ARGV[1]; shift(@ARGV); | |
3046 | if ($arg !~ /^['"]/) { | |
3047 | $matched_pattern = 1; | |
3048 | last; | |
3049 | } | |
3050 | if ($arg =~ s/^(['"])(.*)\1$/$2/) { | |
3051 | $matched_pattern = 1; | |
3052 | last; | |
3053 | } | |
3054 | } | |
3055 | die "Mismatched argument quotes - <$arg>.\n" unless $matched_pattern; | |
3056 | return $arg; | |
3057 | } | |
3058 | ||
d5692f86 SC |
3059 | ####################################################################### |
3060 | # set_worksheet_line($ws_global, $startrow, $startcol, \@content, $format); | |
3061 | # | |
3062 | # set values to a sequence of cells in a row. | |
3063 | # | |
3064 | ####################################################################### | |
3065 | sub set_worksheet_line { | |
3066 | my ($worksheet, $row, $col, $content, $format) = @_; | |
3067 | ||
3068 | foreach my $token (@$content) | |
3069 | { | |
3070 | $worksheet->write($row, $col++, $token, $format ); | |
3071 | } | |
059ec3d9 | 3072 | |
d5692f86 | 3073 | } |
059ec3d9 PH |
3074 | |
3075 | ################################################## | |
3076 | # Main Program # | |
3077 | ################################################## | |
3078 | ||
3079 | ||
3080 | $last_timestamp = ''; | |
3081 | $last_date = ''; | |
3082 | $show_errors = 1; | |
3083 | $show_relay = 1; | |
3084 | $show_transport = 1; | |
3085 | $topcount = 50; | |
3086 | $local_league_table = 1; | |
3087 | $include_remote_users = 0; | |
d5692f86 | 3088 | $include_original_destination = 0; |
059ec3d9 PH |
3089 | $hist_opt = 1; |
3090 | $volume_rounding = 1; | |
3091 | $localtime_offset = calculate_localtime_offset(); # PH/FANF | |
3092 | ||
3093 | $charts = 0; | |
3094 | $charts_option_specified = 0; | |
3095 | $chartrel = "."; | |
3096 | $chartdir = "."; | |
3097 | ||
3098 | @queue_times = (60, 5*60, 15*60, 30*60, 60*60, 3*60*60, 6*60*60, | |
3099 | 12*60*60, 24*60*60); | |
3100 | ||
3101 | $last_offset = ''; | |
3102 | $offset_seconds = 0; | |
3103 | ||
d5692f86 SC |
3104 | $row=1; |
3105 | $row_league_table=1; | |
3106 | $col=0; | |
3107 | $col_hist=0; | |
3108 | $run_hist=0; | |
3109 | my(%output_files); # What output files have been specified? | |
3110 | ||
059ec3d9 PH |
3111 | # Decode options |
3112 | ||
3113 | while (@ARGV > 0 && substr($ARGV[0], 0, 1) eq '-') | |
3114 | { | |
3115 | if ($ARGV[0] =~ /^\-h(\d+)$/) { $hist_opt = $1 } | |
3116 | elsif ($ARGV[0] =~ /^\-ne$/) { $show_errors = 0 } | |
3117 | elsif ($ARGV[0] =~ /^\-nr(.?)(.*)\1$/) | |
3118 | { | |
3119 | if ($1 eq "") { $show_relay = 0 } else { $relay_pattern = $2 } | |
3120 | } | |
3121 | elsif ($ARGV[0] =~ /^\-q([,\d\+\-\*\/]+)$/) | |
3122 | { | |
3123 | @queue_times = split(/,/, $1); | |
3124 | my($q); | |
3125 | foreach $q (@queue_times) { $q = eval($q) + 0 } | |
3126 | @queue_times = sort { $a <=> $b } @queue_times; | |
3127 | @queue_times = () if ($#queue_times == 0 && $queue_times[0] == 0); | |
3128 | } | |
3129 | elsif ($ARGV[0] =~ /^-nt$/) { $show_transport = 0 } | |
3130 | elsif ($ARGV[0] =~ /^\-nt(.?)(.*)\1$/) | |
3131 | { | |
3132 | if ($1 eq "") { $show_transport = 0 } else { $transport_pattern = $2 } | |
3133 | } | |
3134 | elsif ($ARGV[0] =~ /^-t(\d+)$/) { $topcount = $1 } | |
3135 | elsif ($ARGV[0] =~ /^-tnl$/) { $local_league_table = 0 } | |
d5692f86 SC |
3136 | elsif ($ARGV[0] =~ /^-txt=?(\S*)$/) { $txt_fh = get_filehandle($1,\%output_files) } |
3137 | elsif ($ARGV[0] =~ /^-html=?(\S*)$/) { $htm_fh = get_filehandle($1,\%output_files) } | |
3138 | elsif ($ARGV[0] =~ /^-xls=?(\S*)$/) { | |
3139 | if ($HAVE_Spreadsheet_WriteExcel) { | |
3140 | $xls_fh = get_filehandle($1,\%output_files); | |
3141 | } | |
3142 | else { | |
3143 | warn "WARNING: CPAN Module Spreadsheet::WriteExcel not installed. Obtain from www.cpan.org\n"; | |
3144 | } | |
3145 | } | |
059ec3d9 PH |
3146 | elsif ($ARGV[0] =~ /^-merge$/) { $merge_reports = 1 } |
3147 | elsif ($ARGV[0] =~ /^-charts$/) { | |
3148 | $charts = 1; | |
3149 | warn "WARNING: CPAN Module GD::Graph::pie not installed. Obtain from www.cpan.org\n" unless $HAVE_GD_Graph_pie; | |
3150 | warn "WARNING: CPAN Module GD::Graph::linespoints not installed. Obtain from www.cpan.org\n" unless $HAVE_GD_Graph_linespoints; | |
3151 | } | |
3152 | elsif ($ARGV[0] =~ /^-chartdir$/) { $chartdir = $ARGV[1]; shift; $charts_option_specified = 1; } | |
3153 | elsif ($ARGV[0] =~ /^-chartrel$/) { $chartrel = $ARGV[1]; shift; $charts_option_specified = 1; } | |
d5692f86 SC |
3154 | elsif ($ARGV[0] =~ /^-include_original_destination$/) { $include_original_destination = 1 } |
3155 | elsif ($ARGV[0] =~ /^-cache$/) { } #Not currently used. | |
059ec3d9 PH |
3156 | elsif ($ARGV[0] =~ /^-byhost$/) { $do_sender{Host} = 1 } |
3157 | elsif ($ARGV[0] =~ /^-bydomain$/) { $do_sender{Domain} = 1 } | |
3158 | elsif ($ARGV[0] =~ /^-byemail$/) { $do_sender{Email} = 1 } | |
3159 | elsif ($ARGV[0] =~ /^-byemaildomain$/) { $do_sender{Edomain} = 1 } | |
3160 | elsif ($ARGV[0] =~ /^-byedomain$/) { $do_sender{Edomain} = 1 } | |
3161 | elsif ($ARGV[0] =~ /^-nvr$/) { $volume_rounding = 0 } | |
3162 | elsif ($ARGV[0] =~ /^-d$/) { $debug = 1 } | |
3163 | elsif ($ARGV[0] =~ /^--?h(elp)?$/){ help() } | |
3164 | elsif ($ARGV[0] =~ /^-t_remote_users$/) { $include_remote_users = 1 } | |
3165 | elsif ($ARGV[0] =~ /^-pattern$/) | |
3166 | { | |
3167 | push(@user_descriptions,get_next_arg()); | |
3168 | push(@user_patterns,get_next_arg()); | |
3169 | } | |
3170 | elsif ($ARGV[0] =~ /^-utc$/) | |
3171 | { | |
3172 | # We don't need this value if the log is in UTC. | |
3173 | $localtime_offset = undef; | |
3174 | } | |
3175 | else | |
3176 | { | |
3177 | print STDERR "Eximstats: Unknown or malformed option $ARGV[0]\n"; | |
3178 | help(); | |
3179 | } | |
3180 | shift; | |
3181 | } | |
3182 | ||
d5692f86 SC |
3183 | # keep old default behaviour |
3184 | if (! ($xls_fh or $htm_fh or $txt_fh)) { | |
3185 | $txt_fh = \*STDOUT; | |
3186 | } | |
3187 | ||
059ec3d9 PH |
3188 | # Check that all the charts options are specified. |
3189 | warn "-charts option not specified. Use -help for help.\n" if ($charts_option_specified && ! $charts); | |
3190 | ||
3191 | # Default to display tables by sending Host. | |
3192 | $do_sender{Host} = 1 unless ($do_sender{Domain} || $do_sender{Email} || $do_sender{Edomain}); | |
3193 | ||
d5692f86 SC |
3194 | # prepare xls Excel Workbook |
3195 | if (defined $xls_fh) | |
3196 | { | |
3197 | ||
3198 | # Create a new Excel workbook | |
3199 | $workbook = Spreadsheet::WriteExcel->new($xls_fh); | |
3200 | ||
3201 | # Add worksheets | |
3202 | $ws_global = $workbook->addworksheet('Exim Statistik'); | |
3203 | # show $ws_global as initial sheet | |
3204 | $ws_global->set_first_sheet(); | |
3205 | $ws_global->activate(); | |
3206 | ||
3207 | if ($show_relay) { | |
3208 | $ws_relayed = $workbook->addworksheet('Relayed Messages'); | |
3209 | $ws_relayed->set_column(1, 2, 80); | |
3210 | } | |
3211 | if ($topcount) { | |
3212 | $ws_top50 = $workbook->addworksheet('Deliveries'); | |
3213 | } | |
3214 | if ($show_errors) { | |
3215 | $ws_errors = $workbook->addworksheet('Errors'); | |
3216 | } | |
3217 | ||
3218 | ||
3219 | # set column widths | |
3220 | $ws_global->set_column(0, 2, 20); # Columns B-D width set to 30 | |
3221 | $ws_global->set_column(3, 3, 15); # Columns B-D width set to 30 | |
3222 | $ws_global->set_column(4, 4, 25); # Columns B-D width set to 30 | |
3223 | ||
3224 | # Define Formats | |
3225 | $f_default = $workbook->add_format(); | |
3226 | ||
3227 | $f_header1 = $workbook->add_format(); | |
3228 | $f_header1->set_bold(); | |
3229 | #$f_header1->set_color('red'); | |
3230 | $f_header1->set_size('15'); | |
3231 | $f_header1->set_valign(); | |
3232 | # $f_header1->set_align('center'); | |
3233 | # $ws_global->write($row++, 2, "Testing Headers 1", $f_header1); | |
3234 | ||
3235 | $f_header2 = $workbook->add_format(); | |
3236 | $f_header2->set_bold(); | |
3237 | $f_header2->set_size('12'); | |
3238 | $f_header2->set_valign(); | |
3239 | # $ws_global->write($row++, 2, "Testing Headers 2", $f_header2); | |
3240 | ||
3241 | $f_percent = $workbook->add_format(); | |
3242 | $f_percent->set_num_format('0.0%'); | |
3243 | ||
3244 | $f_headertab = $workbook->add_format(); | |
3245 | $f_headertab->set_bold(); | |
3246 | $f_headertab->set_valign(); | |
3247 | # $ws_global->write($row++, 2, "Testing Headers tab", $f_headertab); | |
3248 | ||
3249 | } | |
3250 | ||
059ec3d9 PH |
3251 | |
3252 | for (my $i = 0; $i <= $#queue_times; $i++) { | |
3253 | $queue_bin[$i] = 0; | |
3254 | $remote_queue_bin[$i] = 0; | |
3255 | } | |
3256 | ||
3257 | # Compute the number of slots for the histogram | |
3258 | ||
3259 | if ($hist_opt > 0) | |
3260 | { | |
3261 | if ($hist_opt > 60 || 60 % $hist_opt != 0) | |
3262 | { | |
d5692f86 | 3263 | print STDERR "Eximstats: -h must specify a factor of 60\n"; |
059ec3d9 PH |
3264 | exit 1; |
3265 | } | |
d5692f86 SC |
3266 | $hist_interval = 60/$hist_opt; #Interval in minutes. |
3267 | $hist_number = (24*60)/$hist_interval; #Number of intervals per day. | |
059ec3d9 PH |
3268 | @received_interval_count = (0) x $hist_number; |
3269 | @delivered_interval_count = (0) x $hist_number; | |
3270 | } | |
3271 | ||
3272 | #$queue_unknown = 0; | |
3273 | ||
3274 | $total_received_data = 0; | |
3275 | $total_received_data_gigs = 0; | |
3276 | $total_received_count = 0; | |
3277 | ||
3278 | $total_delivered_data = 0; | |
3279 | $total_delivered_data_gigs = 0; | |
3280 | $total_delivered_count = 0; | |
3281 | ||
3282 | $queue_more_than = 0; | |
3283 | $delayed_count = 0; | |
3284 | $relayed_unshown = 0; | |
3285 | $begin = "9999-99-99 99:99:99"; | |
3286 | $end = "0000-00-00 00:00:00"; | |
3287 | my($section,$type); | |
3288 | foreach $section ('Received','Delivered') { | |
3289 | foreach $type ('Volume','Messages','Delayed','Failed','Hosts','Domains','Emails','Edomains') { | |
3290 | $report_totals{$section}{$type} = 0; | |
3291 | } | |
3292 | } | |
3293 | ||
3294 | # Generate our parser. | |
3295 | my $parser = generate_parser(); | |
3296 | ||
3297 | ||
3298 | ||
3299 | if (@ARGV) { | |
3300 | # Scan the input files and collect the data | |
3301 | foreach my $file (@ARGV) { | |
3302 | if ($file =~ /\.gz/) { | |
3303 | unless (open(FILE,"gunzip -c $file |")) { | |
d5692f86 SC |
3304 | print STDERR "Failed to gunzip -c $file: $!"; |
3305 | next; | |
059ec3d9 PH |
3306 | } |
3307 | } | |
3308 | elsif ($file =~ /\.Z/) { | |
3309 | unless (open(FILE,"uncompress -c $file |")) { | |
d5692f86 SC |
3310 | print STDERR "Failed to uncompress -c $file: $!"; |
3311 | next; | |
059ec3d9 PH |
3312 | } |
3313 | } | |
3314 | else { | |
3315 | unless (open(FILE,$file)) { | |
d5692f86 SC |
3316 | print STDERR "Failed to read $file: $!"; |
3317 | next; | |
059ec3d9 PH |
3318 | } |
3319 | } | |
3320 | #Now parse the filehandle, updating the global variables. | |
3321 | parse($parser,\*FILE); | |
3322 | close FILE; | |
3323 | } | |
3324 | } | |
3325 | else { | |
3326 | #No files provided. Parse STDIN, updating the global variables. | |
3327 | parse($parser,\*STDIN); | |
3328 | } | |
3329 | ||
3330 | ||
3331 | if ($begin eq "9999-99-99 99:99:99") { | |
d5692f86 | 3332 | print STDERR "**** No valid log lines read\n"; |
059ec3d9 PH |
3333 | exit 1; |
3334 | } | |
3335 | ||
3336 | # Output our results. | |
3337 | print_header(); | |
3338 | print_grandtotals(); | |
3339 | ||
3340 | # Print counts of user specified patterns if required. | |
3341 | print_user_patterns() if @user_patterns; | |
3342 | ||
3343 | # Print totals by transport if required. | |
3344 | print_transport() if $show_transport; | |
3345 | ||
3346 | # Print the deliveries per interval as a histogram, unless configured not to. | |
3347 | # First find the maximum in one interval and scale accordingly. | |
3348 | if ($hist_opt > 0) { | |
3349 | print_histogram("Messages received", @received_interval_count); | |
3350 | print_histogram("Deliveries", @delivered_interval_count); | |
3351 | } | |
3352 | ||
3353 | # Print times on queue if required. | |
3354 | if ($#queue_times >= 0) { | |
3355 | print_queue_times("all messages", \@queue_bin,$queue_more_than); | |
3356 | print_queue_times("messages with at least one remote delivery",\@remote_queue_bin,$queue_more_than); | |
3357 | } | |
3358 | ||
3359 | # Print relay information if required. | |
3360 | print_relay() if $show_relay; | |
3361 | ||
3362 | # Print the league tables, if topcount isn't zero. | |
3363 | if ($topcount > 0) { | |
3364 | foreach ('Host','Domain','Email','Edomain') { | |
3365 | next unless $do_sender{$_}; | |
3366 | print_league_table("sending \l$_", $received_count{$_}, $received_data{$_},$received_data_gigs{$_}); | |
3367 | } | |
3368 | ||
3369 | print_league_table("local sender", \%received_count_user, | |
3370 | \%received_data_user,\%received_data_gigs_user) if ($local_league_table || $include_remote_users); | |
3371 | foreach ('Host','Domain','Email','Edomain') { | |
3372 | next unless $do_sender{$_}; | |
3373 | print_league_table("\l$_ destination", $delivered_count{$_}, $delivered_data{$_},$delivered_data_gigs{$_}); | |
3374 | } | |
3375 | print_league_table("local destination", \%delivered_count_user, | |
3376 | \%delivered_data_user,\%delivered_data_gigs_user) if ($local_league_table || $include_remote_users); | |
3377 | } | |
3378 | ||
3379 | # Print the error statistics if required. | |
3380 | print_errors() if $show_errors; | |
3381 | ||
d5692f86 SC |
3382 | print $htm_fh "</body>\n</html>\n" if $htm_fh; |
3383 | ||
3384 | ||
3385 | $txt_fh->close if $txt_fh; | |
3386 | $htm_fh->close if $htm_fh; | |
3387 | ||
3388 | if ($xls_fh) { | |
3389 | # close Excel Workbook | |
3390 | $ws_global->set_first_sheet(); | |
3391 | # FIXME: whyever - activate does not work :-/ | |
3392 | $ws_global->activate(); | |
3393 | $workbook->close(); | |
059ec3d9 PH |
3394 | } |
3395 | ||
d5692f86 | 3396 | |
059ec3d9 | 3397 | # End of eximstats |
d5692f86 SC |
3398 | |
3399 | # FIXME: Doku |