X-Git-Url: https://code.citadel.org/?a=blobdiff_plain;f=citadel%2Fmodules%2Frssclient%2Fserv_rssclient.c;h=85d7f0c1b2675c8e068520790e610e13b7fc4d36;hb=f3d67d7b31c63a0830c714420f6c3af0fdbbb13b;hp=2b1baa5d9b14f014a6677f3d8e355729bcea1fe0;hpb=83b162e34bdb8474a872ae531b56c4cbb934a9a6;p=citadel.git diff --git a/citadel/modules/rssclient/serv_rssclient.c b/citadel/modules/rssclient/serv_rssclient.c index 2b1baa5d9..85d7f0c1b 100644 --- a/citadel/modules/rssclient/serv_rssclient.c +++ b/citadel/modules/rssclient/serv_rssclient.c @@ -1,14 +1,16 @@ /* - * Bring external RSS feeds into rooms. + * Bring external RSS and/or Atom feeds into rooms. This module implements a + * very loose parser that scrapes both kinds of feeds and is not picky about + * the standards compliance of the source data. * - * Copyright (c) 2007-2017 by the citadel.org team + * Copyright (c) 2007-2020 by the citadel.org team * - * This program is open source software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 3. + * This program is open source software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 3. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.See the + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. */ @@ -46,7 +48,6 @@ #include "parsedate.h" #include "database.h" #include "citadel_dirs.h" -#include "md5.h" #include "context.h" #include "internet_addressing.h" @@ -71,7 +72,6 @@ struct rssparser { }; time_t last_run = 0L; -struct CitContext rss_CC; struct rssurl *rsstodo = NULL; @@ -123,7 +123,6 @@ void rss_end_element(void *data, const char *el) (!strcasecmp(el, "entry")) || (!strcasecmp(el, "item")) ) { - if (r->msg != NULL) { // Save the message to the rooms // use the link as an item id if nothing else is available @@ -134,7 +133,7 @@ void rss_end_element(void *data, const char *el) // check the use table StrBuf *u = NewStrBuf(); StrBufAppendPrintf(u, "rss/%s", r->item_id); - time_t already_seen = CheckIfAlreadySeen("rss", u, time(NULL), 604800, eUpdate, 0, 0); + int already_seen = CheckIfAlreadySeen(u); FreeStrBuf(&u); if (already_seen == 0) { @@ -176,16 +175,16 @@ void rss_end_element(void *data, const char *el) long msgnum = (-1); for (rr=r->rooms; rr!=NULL; rr=rr->next) { if (rr == r->rooms) { - msgnum = CtdlSubmitMsg(r->msg, NULL, rr->room, 0); + msgnum = CtdlSubmitMsg(r->msg, NULL, rr->room, 0); // in first room, save msg } else { - CtdlSaveMsgPointerInRoom(rr->room, msgnum, 0, NULL); + CtdlSaveMsgPointerInRoom(rr->room, msgnum, 0, NULL); // elsewhere, save a pointer } - syslog(LOG_DEBUG, "Saved message %ld to %s", msgnum, rr->room); + syslog(LOG_DEBUG, "rssclient: saved message %ld to %s", msgnum, rr->room); } } else { - syslog(LOG_DEBUG, "%s was already seen %ld seconds ago", r->item_id, already_seen); + syslog(LOG_DEBUG, "rssclient: already seen %s", r->item_id); } CM_Free(r->msg); @@ -205,21 +204,36 @@ void rss_end_element(void *data, const char *el) } } - else if (!strcasecmp(el, "author")) { // author of item (rss and maybe atom) + else if (!strcasecmp(el, "creator")) { // can be used if is not present if ((r->msg != NULL) && (CM_IsEmpty(r->msg, eAuthor))) { CM_SetField(r->msg, eAuthor, ChrPtr(r->CData), StrLength(r->CData)); striplt(r->msg->cm_fields[eAuthor]); } } + else if (!strcasecmp(el, "author")) { // supercedes if both are present + if (r->msg != NULL) { + CM_SetField(r->msg, eAuthor, ChrPtr(r->CData), StrLength(r->CData)); // CM_SetField will free() the previous value + striplt(r->msg->cm_fields[eAuthor]); + } + } + else if (!strcasecmp(el, "pubdate")) { // date/time stamp (rss) Sat, 25 Feb 2017 14:28:01 EST - if (CM_IsEmpty(r->msg, eTimestamp)) { + if ((r->msg)&&(r->msg->cm_fields[eTimestamp]==NULL)) { CM_SetFieldLONG(r->msg, eTimestamp, parsedate(ChrPtr(r->CData))); } } else if (!strcasecmp(el, "updated")) { // date/time stamp (atom) 2003-12-13T18:30:02Z - // FIXME parse it + if ((r->msg)&&(r->msg->cm_fields[eTimestamp]==NULL)) { + struct tm t; + char zulu; + memset(&t, 0, sizeof t); + sscanf(ChrPtr(r->CData), "%d-%d-%dT%d:%d:%d%c", &t.tm_year, &t.tm_mon, &t.tm_mday, &t.tm_hour, &t.tm_min, &t.tm_sec, &zulu); + t.tm_year -= 1900; + t.tm_mon -= 1; + CM_SetFieldLONG(r->msg, eTimestamp, mktime(&t)); + } } else if (!strcasecmp(el, "link")) { // link to story (rss) @@ -246,6 +260,7 @@ void rss_end_element(void *data, const char *el) else if ( (!strcasecmp(el, "description")) // message text (rss) || (!strcasecmp(el, "summary")) // message text (atom) + || (!strcasecmp(el, "content")) // message text (atom) ) { if (r->description != NULL) { free(r->description); @@ -284,7 +299,7 @@ void rss_parse_feed(StrBuf *Feed, struct rssroom *rooms) memset(&r, 0, sizeof r); r.rooms = rooms; - XML_Parser p = XML_ParserCreateNS("UTF-8", ':'); + XML_Parser p = XML_ParserCreate("UTF-8"); XML_SetElementHandler(p, rss_start_element, rss_end_element); XML_SetCharacterDataHandler(p, rss_handle_data); XML_SetUserData(p, (void *)&r); @@ -301,13 +316,14 @@ void rssclient_push_todo(char *rssurl, char *roomname) struct rssurl *thisone = NULL; struct rssroom *newroom = NULL; - syslog(LOG_DEBUG, "rssclient_push_todo(%s, %s)", rssurl, roomname); + syslog(LOG_DEBUG, "rssclient: will fetch %s to %s", rssurl, roomname); for (r=rsstodo; r!=NULL; r=r->next) { if (!strcasecmp(r->url, rssurl)) { thisone = r; } } + if (thisone == NULL) { thisone = malloc(sizeof(struct rssurl)); thisone->url = strdup(rssurl); @@ -323,17 +339,6 @@ void rssclient_push_todo(char *rssurl, char *roomname) } -// Callback function for curl -// -size_t rss_pof_write_data(void *buffer, size_t size, size_t nmemb, void *userp) -{ - StrBuf *Downloaded = (StrBuf *)userp; - size_t bytes = size * nmemb; - StrBufAppendBufPlain(Downloaded, buffer, bytes, 0); - return(bytes); -} - - // pull one feed (possibly multiple rooms) // void rss_pull_one_feed(struct rssurl *url) @@ -342,7 +347,7 @@ void rss_pull_one_feed(struct rssurl *url) CURLcode res; StrBuf *Downloaded = NULL; - syslog(LOG_DEBUG, "rss_pull_one_feed(%s)", url->url); + syslog(LOG_DEBUG, "rssclient: fetching %s", url->url); curl = curl_easy_init(); if (!curl) { @@ -352,13 +357,15 @@ void rss_pull_one_feed(struct rssurl *url) Downloaded = NewStrBuf(); curl_easy_setopt(curl, CURLOPT_URL, url->url); + curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, 0L); + curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, 0L); curl_easy_setopt(curl, CURLOPT_FOLLOWLOCATION, 1L); // Follow redirects - curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, rss_pof_write_data); // What to do with downloaded data + curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, CurlFillStrBuf_callback); // What to do with downloaded data curl_easy_setopt(curl, CURLOPT_WRITEDATA, Downloaded); // Give it our StrBuf to work with curl_easy_setopt(curl, CURLOPT_TIMEOUT, 20L); // Time out after 20 seconds res = curl_easy_perform(curl); // Perform the request if (res != CURLE_OK) { - syslog(LOG_WARNING, "Failed to load feed: %s", curl_easy_strerror(res)); + syslog(LOG_WARNING, "rssclient: failed to load feed: %s", curl_easy_strerror(res)); } curl_easy_cleanup(curl); @@ -430,18 +437,17 @@ void rssclient_scan(void) { /* Run no more than once every 15 minutes. */ if ((now - last_run) < 900) { syslog(LOG_DEBUG, - "Client: polling interval not yet reached; last run was %ldm%lds ago", + "rssclient: polling interval not yet reached; last run was %ldm%lds ago", ((now - last_run) / 60), ((now - last_run) % 60) ); return; } - become_session(&rss_CC); - syslog(LOG_DEBUG, "rssclient started"); + syslog(LOG_DEBUG, "rssclient: started"); CtdlForEachRoom(rssclient_scan_room, NULL); rss_pull_feeds(); - syslog(LOG_DEBUG, "rssclient ended"); + syslog(LOG_DEBUG, "rssclient: ended"); last_run = time(NULL); return; } @@ -451,13 +457,9 @@ CTDL_MODULE_INIT(rssclient) { if (!threading) { - syslog(LOG_INFO, "%s", curl_version()); + syslog(LOG_INFO, "rssclient: using %s", curl_version()); CtdlRegisterSessionHook(rssclient_scan, EVT_TIMER, PRIO_AGGR + 300); } - else - { - CtdlFillSystemContext(&rss_CC, "rssclient"); - } return "rssclient"; }