From 29f89cad0cf7be1977f6ed36d27ac9b651aec9e2 Mon Sep 17 00:00:00 2001 From: Philip Hazel Date: Wed, 14 Feb 2007 14:59:01 +0000 Subject: [PATCH] Added expansion items addresses, filter, map, reduce. --- doc/doc-txt/ChangeLog | 5 +- doc/doc-txt/NewStuff | 75 ++++++++++- src/src/expand.c | 248 ++++++++++++++++++++++++++++++++++- test/scripts/0000-Basic/0002 | 28 ++++ test/stdout/0002 | 26 ++++ 5 files changed, 379 insertions(+), 3 deletions(-) diff --git a/doc/doc-txt/ChangeLog b/doc/doc-txt/ChangeLog index 146c4fbfa..61b6ac275 100644 --- a/doc/doc-txt/ChangeLog +++ b/doc/doc-txt/ChangeLog @@ -1,4 +1,4 @@ -$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.478 2007/02/14 12:22:36 steve Exp $ +$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.479 2007/02/14 14:59:01 ph10 Exp $ Change log file for Exim from version 4.21 ------------------------------------------- @@ -101,6 +101,9 @@ PH/24 When the log selector "outgoing_port" was set, the port was shown as -1 for deliveries of the second and subsequent messages over the same SMTP connection. +PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and + ${reduce, with only minor "tidies". + SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match. diff --git a/doc/doc-txt/NewStuff b/doc/doc-txt/NewStuff index 04177227d..b70fa5e68 100644 --- a/doc/doc-txt/NewStuff +++ b/doc/doc-txt/NewStuff @@ -1,4 +1,4 @@ -$Cambridge: exim/doc/doc-txt/NewStuff,v 1.140 2007/02/07 12:23:35 ph10 Exp $ +$Cambridge: exim/doc/doc-txt/NewStuff,v 1.141 2007/02/14 14:59:01 ph10 Exp $ New Features in Exim -------------------- @@ -292,6 +292,79 @@ Version 4.67 option; with -I they don't. In both cases it is possible to change the case sensitivity within the pattern using (?i) or (?-i). +14. A number of new features have been added to string expansions to make it + easier to process lists of items, typically addresses. These are as + follows: + + * ${addresses:} + + The string (after expansion) is interpreted as a list of addresses in RFC + 2822 format, such as can be found in a To: or Cc: header line. The + operative address (local-part@domain) is extracted from each item, and the + result of the expansion is a colon-separated list, with appropriate + doubling of colons should any happen to be present in the email addresses. + Syntactically invalid RFC2822 address items are omitted from the output. + + It is possible to specify a character other than colon for the output + separator by starting the string with > followed by the new separator + character. For example: + + ${addresses:>& The Boss , sec@base.ment (dogsbody)} + + expands to "ceo@up.stairs&sec@base.ment". Compare ${address (singular), + which extracts the working address from a single RFC2822 address. + + * ${map{}{}} + + After expansion, is interpreted as a list, colon-separated by + default, but the separator can be changed in the usual way. For each item + in this list, its value is place in $item, and then is expanded + and added to the output as an item in a new list. The separator used for + the output list is the same as the one used for the input, but is not + included in the output. For example: + + ${map{a:b:c}{[$item]}} ${map{<- x-y-z}{($item)}} + + expands to "[a]:[b]:[c] (x)-(y)-(z)". At the end of the expansion, the + value of $item is restored to what it was before. + + * ${filter{}{}} + + After expansion, is interpreted as a list, colon-separated by + default, but the separator can be changed in the usual way. For each item + in this list, its value is place in $item, and then the condition is + evaluated. If the condition is true, $item is added to the output as an + item in a new list; if the condition is false, the item is discarded. The + separator used for the output list is the same as the one used for the + input, but is not included in the output. For example: + + ${filter{a:b:c}{!eq{$item}{b}} + + yields "a:c". At the end of the expansion, the value of $item is restored + to what it was before. + + * ${reduce{}{}{}} + + The ${reduce expansion operation reduces a list to a single, scalar string. + After expansion, is interpreted as a list, colon-separated by + default, but the separator can be changed in the usual way. Then + is expanded and assigned to the $value variable. After this, each item in + the list is assigned to $item in turn, and is expanded + for each of them. The result of that expansion is assigned to $value before + the next iteration. When the end of the list is reached, the final value of + $value is added to the expansion string. The ${reduce expansion item can be + used in a number of ways. For example, to add up a list of numbers: + + ${reduce {<, 1,2,3}{0}{${eval:$value+$item}}} + + The result of that expansion would be "6". The maximum of a list of numbers + can be found: + + ${reduce {3:0:9:4:6}{0}{${if >{$item}{$value}{$item}{$value}}}} + + At the end of a ${reduce expansion, the values of $item and $value is + restored to what they were before. + Version 4.66 ------------ diff --git a/src/src/expand.c b/src/src/expand.c index 155f4d836..afa898205 100644 --- a/src/src/expand.c +++ b/src/src/expand.c @@ -1,4 +1,4 @@ -/* $Cambridge: exim/src/src/expand.c,v 1.81 2007/02/10 23:51:11 magnus Exp $ */ +/* $Cambridge: exim/src/src/expand.c,v 1.82 2007/02/14 14:59:02 ph10 Exp $ */ /************************************************* * Exim - an Internet mail transport agent * @@ -106,17 +106,20 @@ alphabetical order. */ static uschar *item_table[] = { US"dlfunc", US"extract", + US"filter", US"hash", US"hmac", US"if", US"length", US"lookup", + US"map", US"nhash", US"perl", US"prvs", US"prvscheck", US"readfile", US"readsocket", + US"reduce", US"run", US"sg", US"substr", @@ -125,17 +128,20 @@ static uschar *item_table[] = { enum { EITEM_DLFUNC, EITEM_EXTRACT, + EITEM_FILTER, EITEM_HASH, EITEM_HMAC, EITEM_IF, EITEM_LENGTH, EITEM_LOOKUP, + EITEM_MAP, EITEM_NHASH, EITEM_PERL, EITEM_PRVS, EITEM_PRVSCHECK, EITEM_READFILE, EITEM_READSOCK, + EITEM_REDUCE, EITEM_RUN, EITEM_SG, EITEM_SUBSTR, @@ -162,6 +168,7 @@ enum { static uschar *op_table_main[] = { US"address", + US"addresses", US"base62", US"base62d", US"domain", @@ -193,6 +200,7 @@ static uschar *op_table_main[] = { enum { EOP_ADDRESS = sizeof(op_table_underscore)/sizeof(uschar *), + EOP_ADDRESSES, EOP_BASE62, EOP_BASE62D, EOP_DOMAIN, @@ -4682,6 +4690,181 @@ while (*s != 0) } + /* Handle list operations */ + + case EITEM_FILTER: + case EITEM_MAP: + case EITEM_REDUCE: + { + int sep = 0; + int save_ptr = ptr; + uschar outsep[2] = { '\0', '\0' }; + uschar *list, *expr, *temp; + uschar *save_iterate_item = iterate_item; + uschar *save_lookup_value = lookup_value; + + while (isspace(*s)) s++; + if (*s++ != '{') goto EXPAND_FAILED_CURLY; + + list = expand_string_internal(s, TRUE, &s, skipping); + if (list == NULL) goto EXPAND_FAILED; + if (*s++ != '}') goto EXPAND_FAILED_CURLY; + + if (item_type == EITEM_REDUCE) + { + while (isspace(*s)) s++; + if (*s++ != '{') goto EXPAND_FAILED_CURLY; + temp = expand_string_internal(s, TRUE, &s, skipping); + if (temp == NULL) goto EXPAND_FAILED; + lookup_value = temp; + if (*s++ != '}') goto EXPAND_FAILED_CURLY; + } + + while (isspace(*s)) s++; + if (*s++ != '{') goto EXPAND_FAILED_CURLY; + + expr = s; + + /* For EITEM_FILTER, call eval_condition once, with result discarded (as + if scanning a "false" part). This allows us to find the end of the + condition, because if the list is empty, we won't actually evaluate the + condition for real. For EITEM_MAP and EITEM_REDUCE, do the same, using + the normal internal expansion function. */ + + if (item_type == EITEM_FILTER) + { + temp = eval_condition(expr, NULL); + if (temp != NULL) s = temp; + } + else + { + temp = expand_string_internal(s, TRUE, &s, TRUE); + } + + if (temp == NULL) + { + expand_string_message = string_sprintf("%s inside \"%s\" item", + expand_string_message, name); + goto EXPAND_FAILED; + } + + while (isspace(*s)) s++; + if (*s++ != '}') + { + expand_string_message = string_sprintf("missing } at end of condition " + "or expression inside \"%s\"", name); + goto EXPAND_FAILED; + } + + while (isspace(*s)) s++; + if (*s++ != '}') + { + expand_string_message = string_sprintf("missing } at end of \"%s\"", + name); + goto EXPAND_FAILED; + } + + /* If we are skipping, we can now just move on to the next item. When + processing for real, we perform the iteration. */ + + if (skipping) continue; + while ((iterate_item = string_nextinlist(&list, &sep, NULL, 0)) != NULL) + { + *outsep = (uschar)sep; /* Separator as a string */ + + DEBUG(D_expand) debug_printf("%s: $item = \"%s\"\n", name, iterate_item); + + if (item_type == EITEM_FILTER) + { + BOOL condresult; + if (eval_condition(expr, &condresult) == NULL) + { + expand_string_message = string_sprintf("%s inside \"%s\" condition", + expand_string_message, name); + goto EXPAND_FAILED; + } + DEBUG(D_expand) debug_printf("%s: condition is %s\n", name, + condresult? "true":"false"); + if (condresult) + temp = iterate_item; /* TRUE => include this item */ + else + continue; /* FALSE => skip this item */ + } + + /* EITEM_MAP and EITEM_REDUCE */ + + else + { + temp = expand_string_internal(expr, TRUE, NULL, skipping); + if (temp == NULL) + { + expand_string_message = string_sprintf("%s inside \"%s\" item", + expand_string_message, name); + goto EXPAND_FAILED; + } + if (item_type == EITEM_REDUCE) + { + lookup_value = temp; /* Update the value of $value */ + continue; /* and continue the iteration */ + } + } + + /* We reach here for FILTER if the condition is true, always for MAP, + and never for REDUCE. The value in "temp" is to be added to the output + list that is being created, ensuring that any occurrences of the + separator character are doubled. Unless we are dealing with the first + item of the output list, add in a space if the new item begins with the + separator character, or is an empty string. */ + + if (ptr != save_ptr && (temp[0] == *outsep || temp[0] == 0)) + yield = string_cat(yield, &size, &ptr, US" ", 1); + + /* Add the string in "temp" to the output list that we are building, + This is done in chunks by searching for the separator character. */ + + for (;;) + { + size_t seglen = Ustrcspn(temp, outsep); + yield = string_cat(yield, &size, &ptr, temp, seglen + 1); + + /* If we got to the end of the string we output one character + too many; backup and end the loop. Otherwise arrange to double the + separator. */ + + if (temp[seglen] == '\0') { ptr--; break; } + yield = string_cat(yield, &size, &ptr, outsep, 1); + temp += seglen + 1; + } + + /* Output a separator after the string: we will remove the redundant + final one at the end. */ + + yield = string_cat(yield, &size, &ptr, outsep, 1); + } /* End of iteration over the list loop */ + + /* REDUCE has generated no output above: output the final value of + $value. */ + + if (item_type == EITEM_REDUCE) + { + yield = string_cat(yield, &size, &ptr, lookup_value, + Ustrlen(lookup_value)); + lookup_value = save_lookup_value; /* Restore $value */ + } + + /* FILTER and MAP generate lists: if they have generated anything, remove + the redundant final separator. Even though an empty item at the end of a + list does not count, this is tidier. */ + + else if (ptr != save_ptr) ptr--; + + /* Restore preserved $item */ + + iterate_item = save_iterate_item; + continue; + } + + /* If ${dlfunc support is configured, handle calling dynamically-loaded functions, unless locked out at this time. Syntax is ${dlfunc{file}{func}} or ${dlfunc{file}{func}{arg}} or ${dlfunc{file}{func}{arg1}{arg2}} or up to @@ -5042,6 +5225,69 @@ while (*s != 0) continue; } + case EOP_ADDRESSES: + { + uschar outsep[2] = { ':', '\0' }; + uschar *address, *error; + int save_ptr = ptr; + int start, end, domain; /* Not really used */ + + while (isspace(*sub)) sub++; + if (*sub == '>') { *outsep = *++sub; ++sub; } + parse_allow_group = TRUE; + + for (;;) + { + uschar *p = parse_find_address_end(sub, FALSE); + uschar saveend = *p; + *p = '\0'; + address = parse_extract_address(sub, &error, &start, &end, &domain, + FALSE); + *p = saveend; + + /* Add the address to the output list that we are building. This is + done in chunks by searching for the separator character. At the + start, unless we are dealing with the first address of the output + list, add in a space if the new address begins with the separator + character, or is an empty string. */ + + if (address != NULL) + { + if (ptr != save_ptr && address[0] == *outsep) + yield = string_cat(yield, &size, &ptr, US" ", 1); + + for (;;) + { + size_t seglen = Ustrcspn(address, outsep); + yield = string_cat(yield, &size, &ptr, address, seglen + 1); + + /* If we got to the end of the string we output one character + too many. */ + + if (address[seglen] == '\0') { ptr--; break; } + yield = string_cat(yield, &size, &ptr, outsep, 1); + address += seglen + 1; + } + + /* Output a separator after the string: we will remove the + redundant final one at the end. */ + + yield = string_cat(yield, &size, &ptr, outsep, 1); + } + + if (saveend == '\0') break; + sub = p + 1; + } + + /* If we have generated anything, remove the redundant final + separator. */ + + if (ptr != save_ptr) ptr--; + parse_allow_group = FALSE; + continue; + } + + /* quote puts a string in quotes if it is empty or contains anything other than alphamerics, underscore, dot, or hyphen. diff --git a/test/scripts/0000-Basic/0002 b/test/scripts/0000-Basic/0002 index 5e0126524..870359eb2 100644 --- a/test/scripts/0000-Basic/0002 +++ b/test/scripts/0000-Basic/0002 @@ -40,12 +40,40 @@ x\ +$11111111111111111111111111111111111 +${11111111111111111111111111111111111} +# List operations + +filter: "${filter{a:b:c}{eq{1}{1}}}" +filter: ${filter{a:b:c}{!eq{$item}{b}}} +filter: ${filter{<' a'b'c}{!eq{$item}{b}}} +filter: ${filter{<' ''a'b' ''c}{!eq{$item}{b}}} +filter: "${filter{}{!eq{$item}{b}}}" + +map: "${map{}{$item}}" +map: ${map{a:b:c}{$item}} +map: ${map{a:b:c}{:$item:}} +map: ${if eq{1}{0}{${map{a:b:c}{:$item:}}}{fail string}} +map: ${map{:b:c}{[$item]}} + +reduce: "${reduce{}{+}{$value$item}}" +reduce: ${reduce{a:b:c}{+}{$value$item}} +reduce: ${reduce {<, 1,2,3}{0}{${eval:$value+$item}}} +reduce: ${reduce {3:0:9:4:6}{0}{${if >{$item}{$value}{$item}{$value}}}} + # Operators addrss: ${address:local-part@dom.ain} addrss: ${address:Exim Person (that's me)} domain: ${domain:local-part@dom.ain} domain: ${domain:Exim Person (that's me)} + +addresses: ${addresses:>' 'abc@xyz, 'pqr@xyz} +addresses: ${addresses:Exim Person (that's me)} +addresses: ${addresses:>+ Exim Person (that's me),\ + xyz@abc} +addresses: ${addresses:Exim Person (that's me), \ + xyz@abc, nullgroupname:;, group: p@q, r@s; } +addresses: ${addresses:local-part@dom.ain } + escape: ${escape:B7·F2ò} eval: ${eval:1+1} eval: ${eval:1+2*3} diff --git a/test/stdout/0002 b/test/stdout/0002 index b1f29c447..d985a9c1e 100644 --- a/test/stdout/0002 +++ b/test/stdout/0002 @@ -30,12 +30,38 @@ > + > + > +> # List operations +> +> filter: "a:b:c" +> filter: a:c +> filter: a'c +> filter: ''a' ''c +> filter: "" +> +> map: "" +> map: a:b:c +> map: ::a::: ::b::: ::c:: +> map: fail string +> map: []:[b]:[c] +> +> reduce: "+" +> reduce: +abc +> reduce: 6 +> reduce: 9 +> > # Operators > > addrss: local-part@dom.ain > addrss: local-part@dom.ain > domain: dom.ain > domain: dom.ain +> +> addresses: ''abc@xyz' ''pqr@xyz +> addresses: local-part@dom.ain +> addresses: local-part@dom.ain+xyz@abc +> addresses: local-part@dom.ain:xyz@abc:p@q:r@s +> addresses: +> > escape: B7\267F2\362 > eval: 2 > eval: 7 -- 2.25.1