Index: modules/metadata/mod_setenvif.c =================================================================== --- modules/metadata/mod_setenvif.c (revision 410296) +++ modules/metadata/mod_setenvif.c (working copy) @@ -102,7 +102,8 @@ SPECIAL_REQUEST_URI, SPECIAL_REQUEST_METHOD, SPECIAL_REQUEST_PROTOCOL, - SPECIAL_SERVER_ADDR + SPECIAL_SERVER_ADDR, + SPECIAL_OUTPUT_HEADER }; typedef struct { char *name; /* header name */ @@ -113,10 +114,12 @@ apr_table_t *features; /* env vars to set (or unset) */ enum special special_type; /* is it a "special" header ? */ int icase; /* ignoring case? */ + int iscontent_type; /* is content type ? */ } sei_entry; typedef struct { apr_array_header_t *conditionals; + int output_filter_on; /* act as output_filter for headers_out ? */ } sei_cfg_rec; module AP_MODULE_DECLARE_DATA setenvif_module; @@ -135,6 +138,7 @@ sei_cfg_rec *new = (sei_cfg_rec *) apr_palloc(p, sizeof(sei_cfg_rec)); new->conditionals = apr_array_make(p, 20, sizeof(sei_entry)); + new->output_filter_on = 0; return (void *) new; } @@ -155,6 +159,8 @@ a->conditionals = apr_array_append(p, base->conditionals, overrides->conditionals); + if(base->output_filter_on || overrides->output_filter_on) + a->output_filter_on = 1; return a; } @@ -164,6 +170,7 @@ */ #define ICASE_MAGIC ((void *)(&setenvif_module)) #define SEI_MAGIC_HEIRLOOM "setenvif-phase-flag" +#define SEI_OUTPUT_FILTER_NAME "setenvif-output-filter" static int is_header_regex(apr_pool_t *p, const char* name) { @@ -307,26 +314,7 @@ new->name = fname; new->regex = regex; new->icase = icase; - if ((simple_pattern = non_regex_pattern(cmd->pool, regex))) { - new->pattern = apr_strmatch_precompile(cmd->pool, - simple_pattern, !icase); - if (new->pattern == NULL) { - return apr_pstrcat(cmd->pool, cmd->cmd->name, - " pattern could not be compiled.", NULL); - } - new->preg = NULL; - } - else { - new->preg = ap_pregcomp(cmd->pool, regex, - (REG_EXTENDED | (icase ? REG_ICASE : 0))); - if (new->preg == NULL) { - return apr_pstrcat(cmd->pool, cmd->cmd->name, - " regex could not be compiled.", NULL); - } - new->pattern = NULL; - } - new->features = apr_table_make(cmd->pool, 2); - + new->iscontent_type = 0; if (!strcasecmp(fname, "remote_addr")) { new->special_type = SPECIAL_REMOTE_ADDR; } @@ -345,7 +333,37 @@ else if (!strcasecmp(fname, "server_addr")) { new->special_type = SPECIAL_SERVER_ADDR; } - else { + else if (!strcasecmp(fname, "output_header")) { + new->special_type = SPECIAL_OUTPUT_HEADER; + sconf->output_filter_on = 1; + /* + * The regex is the 3rd word here, because we keep the same syntax + * as for header string, but we added the output_header key word to + * keep backward compatibility. (to avoid the matching of both + * request and response header on a regex). + * In this case the 2nd param of the directive is the header + * regex/name. + */ + new->name = regex; + if (is_header_regex(cmd->pool, new->name)) { + new->pnamereg = ap_pregcomp(cmd->pool, new->name, (REG_EXTENDED | + REG_NOSUB | (icase ? REG_ICASE : 0))); + if (new->pnamereg == NULL) + return apr_pstrcat(cmd->pool, cmd->cmd->name, "Header name regex could not be compiled.", NULL); + } + else { + new->pnamereg = NULL; + if(!strcasecmp(new->name, "content-type")) + new->iscontent_type = 1; + } + regex = ap_getword_conf(cmd->pool, &args); + if (!*regex) { + return apr_pstrcat(cmd->pool, "Missing regular expression for ", + cmd->cmd->name, NULL); + } + new->regex = regex; + } + else { new->special_type = SPECIAL_NOT; /* Handle fname as a regular expression. * If fname a simple header string, identify as such @@ -365,6 +383,30 @@ new->pnamereg = NULL; } } + /* + * This block must be after the identification of the parameters, as + * the regex may be in an other position than the second one depends + * of the special type. + */ + if ((simple_pattern = non_regex_pattern(cmd->pool, regex))) { + new->pattern = apr_strmatch_precompile(cmd->pool, + simple_pattern, !icase); + if (new->pattern == NULL) { + return apr_pstrcat(cmd->pool, cmd->cmd->name, + " pattern could not be compiled.", NULL); + } + new->preg = NULL; + } + else { + new->preg = ap_pregcomp(cmd->pool, regex, + (REG_EXTENDED | (icase ? REG_ICASE : 0))); + if (new->preg == NULL) { + return apr_pstrcat(cmd->pool, cmd->cmd->name, + " regex could not be compiled.", NULL); + } + new->pattern = NULL; + } + new->features = apr_table_make(cmd->pool, 2); } else { new = &entries[i]; @@ -436,6 +478,95 @@ }; /* + * This routine get from a header (or subprocess_env if no match in header), + * the value that correspond to a header name or a header matching regexp. + * (according to a sei_entry). + */ +static void get_header_val(sei_entry *b, apr_table_t *headers, apr_table_t *subprocess_env, char const **val, request_rec *r) { + const apr_table_entry_t *elts; + /* Matching headers_in against a regex. Iterate through + * the headers until we find a match or run out of + * headers. + */ + const apr_array_header_t *arr = apr_table_elts(headers); + int j; + + + if (b->pnamereg) { + elts = (const apr_table_entry_t *) arr->elts; + *val = NULL; + for (j = 0; j < arr->nelts; ++j) { + if (!ap_regexec(b->pnamereg, elts[j].key, 0, NULL, 0)) { + *val = elts[j].val; + } + } + } + else { + if(!b->iscontent_type) { + /* Not matching against a regex */ + *val = apr_table_get(headers, b->name); + if (*val == NULL) { + *val = apr_table_get(subprocess_env, b->name); + } + } + else { + *val = r->content_type; + } + } +} + + +/* + * This routine set environment variable according to val. + * Question : why sei_entry's not const ? because ap_regexec is not defined as + * well... pcreposix does not define the use of the regex as not modifying it? + */ +static void setenvif_val_match(request_rec *r, const char *val, apr_size_t + val_len, sei_entry *b) { + regmatch_t regm[AP_MAX_REG_MATCH]; + const apr_table_entry_t *elts; + int j; + + /* + * A NULL value indicates that the header field or special entity + * wasn't present or is undefined. Represent that as an empty string + * so that REs like "^$" will work and allow envariable setting + * based on missing or empty field. + */ + if (val == NULL) { + val = ""; + val_len = 0; + } + + if ((b->pattern && apr_strmatch(b->pattern, val, val_len)) || + (!b->pattern && !ap_regexec(b->preg, val, AP_MAX_REG_MATCH, regm, + 0))) { + const apr_array_header_t *arr = apr_table_elts(b->features); + elts = (const apr_table_entry_t *) arr->elts; + + for (j = 0; j < arr->nelts; ++j) { + if (*(elts[j].val) == '!') { + apr_table_unset(r->subprocess_env, elts[j].key); + } + else { + if (!b->pattern) { + char *replaced = ap_pregsub(r->pool, elts[j].val, val, + AP_MAX_REG_MATCH, regm); + if (replaced) { + apr_table_setn(r->subprocess_env, elts[j].key, + replaced); + } + } + else { + apr_table_setn(r->subprocess_env, elts[j].key, + elts[j].val); + } + } + } + } +} + +/* * This routine gets called at two different points in request processing: * once before the URI has been translated (during the post-read-request * phase) and once after (during the header-parse phase). We use different @@ -448,12 +579,10 @@ { sei_cfg_rec *sconf; sei_entry *entries; - const apr_table_entry_t *elts; const char *val; apr_size_t val_len = 0; - int i, j; + int i; char *last_name; - regmatch_t regm[AP_MAX_REG_MATCH]; if (!ap_get_module_config(r->request_config, &setenvif_module)) { ap_set_module_config(r->request_config, &setenvif_module, @@ -467,10 +596,11 @@ } entries = (sei_entry *) sconf->conditionals->elts; last_name = NULL; - val = NULL; + for (i = 0; i < sconf->conditionals->nelts; ++i) { sei_entry *b = &entries[i]; + val = NULL; /* Optimize the case where a bunch of directives in a row use the * same header. Remember we don't need to strcmp the two header * names because we made sure the pointers were equal during @@ -499,79 +629,73 @@ val = r->protocol; break; case SPECIAL_NOT: - if (b->pnamereg) { - /* Matching headers_in against a regex. Iterate through - * the headers_in until we find a match or run out of - * headers. - */ - const apr_array_header_t - *arr = apr_table_elts(r->headers_in); - - elts = (const apr_table_entry_t *) arr->elts; - val = NULL; - for (j = 0; j < arr->nelts; ++j) { - if (!ap_regexec(b->pnamereg, elts[j].key, 0, NULL, 0)) { - val = elts[j].val; - } - } - } - else { - /* Not matching against a regex */ - val = apr_table_get(r->headers_in, b->name); - if (val == NULL) { - val = apr_table_get(r->subprocess_env, b->name); - } - } + get_header_val(b, r->headers_in, r->subprocess_env, &val, r); } val_len = val ? strlen(val) : 0; } + setenvif_val_match(r, val, val_len, b); + + } + + if (sconf->output_filter_on == 1) + ap_add_output_filter(SEI_OUTPUT_FILTER_NAME, sconf, r, r->connection); + + return DECLINED; +} + +/* + * This (output) filter process headers_out to find condition to set env. + */ +static apr_status_t setenvif_out_filter(ap_filter_t * f, apr_bucket_brigade * bb) +{ + sei_cfg_rec *sconf; + sei_entry *entries; + const char *val; + apr_size_t val_len = 0; + int i; + char *last_name; + + sconf = (sei_cfg_rec *) f->ctx; + entries = (sei_entry *) sconf->conditionals->elts; + last_name = NULL; + for (i = 0; i < sconf->conditionals->nelts; ++i) { + sei_entry *b = &entries[i]; + + val = NULL; /* - * A NULL value indicates that the header field or special entity - * wasn't present or is undefined. Represent that as an empty string - * so that REs like "^$" will work and allow envariable setting - * based on missing or empty field. + * Optimize the case where a bunch of directives in a row use the + * same header. Remember we don't need to strcmp the two header + * names because we made sure the pointers were equal during + * configuration. */ - if (val == NULL) { - val = ""; - val_len = 0; - } - - if ((b->pattern && apr_strmatch(b->pattern, val, val_len)) || - (!b->pattern && !ap_regexec(b->preg, val, AP_MAX_REG_MATCH, regm, - 0))) { - const apr_array_header_t *arr = apr_table_elts(b->features); - elts = (const apr_table_entry_t *) arr->elts; - - for (j = 0; j < arr->nelts; ++j) { - if (*(elts[j].val) == '!') { - apr_table_unset(r->subprocess_env, elts[j].key); - } - else { - if (!b->pattern) { - char *replaced = ap_pregsub(r->pool, elts[j].val, val, - AP_MAX_REG_MATCH, regm); - if (replaced) { - apr_table_setn(r->subprocess_env, elts[j].key, - replaced); - } - } - else { - apr_table_setn(r->subprocess_env, elts[j].key, - elts[j].val); - } - } + if (b->name != last_name) { + last_name = b->name; + if (b->special_type == SPECIAL_OUTPUT_HEADER) { + /* + * As output filter, only headers_out are interresting for us + * so we just process this type of conditions. + */ + get_header_val(b, f->r->headers_out, f->r->subprocess_env, &val, f->r); + val_len = val ? strlen(val) : 0; + setenvif_val_match(f->r, val, val_len, b); } } } - return DECLINED; + ap_remove_output_filter(f); + return ap_pass_brigade(f->next, bb); } static void register_hooks(apr_pool_t *p) { ap_hook_header_parser(match_headers, NULL, NULL, APR_HOOK_MIDDLE); ap_hook_post_read_request(match_headers, NULL, NULL, APR_HOOK_MIDDLE); + /* + * ftype is set to AP_FTYPE_CONTENT_SET + 1 to make this filter hooked + * before deflate (it's specific for the moment). + */ + ap_register_output_filter(SEI_OUTPUT_FILTER_NAME, setenvif_out_filter, NULL, AP_FTYPE_CONTENT_SET); } module AP_MODULE_DECLARE_DATA setenvif_module =