Add generic filter/plugin infrastructure

The functions cgit_open_filter() and cgit_close_filter() can be used to
execute filters on the output stream from cgit.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>

3 files changed, 62 insertions, 0 deletions

diff --git a/cgit.c b/cgit.c
index 2039ab1..779a464 100644
--- a/cgit.c
+++ b/cgit.c
@@ -1,147 +1,162 @@
 /* cgit.c: cgi for the git scm
  *
  * Copyright (C) 2006 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
 #include "cgit.h"
 #include "cache.h"
 #include "cmd.h"
 #include "configfile.h"
 #include "html.h"
 #include "ui-shared.h"
 #include "ui-stats.h"
 #include "scan-tree.h"
 
 const char *cgit_version = CGIT_VERSION;
 
+struct cgit_filter *new_filter(const char *cmd, int extra_args)
+{
+        struct cgit_filter *f;
+
+        if (!cmd)
+                return NULL;
+
+        f = xmalloc(sizeof(struct cgit_filter));
+        f->cmd = xstrdup(cmd);
+        f->argv = xmalloc((2 + extra_args) * sizeof(char *));
+        f->argv[0] = f->cmd;
+        f->argv[1] = NULL;
+        return f;
+}
+
 void config_cb(const char *name, const char *value)
 {
         if (!strcmp(name, "root-title"))
                 ctx.cfg.root_title = xstrdup(value);
         else if (!strcmp(name, "root-desc"))
                 ctx.cfg.root_desc = xstrdup(value);
         else if (!strcmp(name, "root-readme"))
                 ctx.cfg.root_readme = xstrdup(value);
         else if (!strcmp(name, "css"))
                 ctx.cfg.css = xstrdup(value);
         else if (!strcmp(name, "favicon"))
                 ctx.cfg.favicon = xstrdup(value);
         else if (!strcmp(name, "footer"))
                 ctx.cfg.footer = xstrdup(value);
         else if (!strcmp(name, "head-include"))
                 ctx.cfg.head_include = xstrdup(value);
         else if (!strcmp(name, "header"))
                 ctx.cfg.header = xstrdup(value);
         else if (!strcmp(name, "logo"))
                 ctx.cfg.logo = xstrdup(value);
         else if (!strcmp(name, "index-header"))
                 ctx.cfg.index_header = xstrdup(value);
         else if (!strcmp(name, "index-info"))
                 ctx.cfg.index_info = xstrdup(value);
         else if (!strcmp(name, "logo-link"))
                 ctx.cfg.logo_link = xstrdup(value);
         else if (!strcmp(name, "module-link"))
                 ctx.cfg.module_link = xstrdup(value);
         else if (!strcmp(name, "virtual-root")) {
                 ctx.cfg.virtual_root = trim_end(value, '/');
                 if (!ctx.cfg.virtual_root && (!strcmp(value, "/")))
                         ctx.cfg.virtual_root = "";
         } else if (!strcmp(name, "nocache"))
                 ctx.cfg.nocache = atoi(value);
         else if (!strcmp(name, "noheader"))
                 ctx.cfg.noheader = atoi(value);
         else if (!strcmp(name, "snapshots"))
                 ctx.cfg.snapshots = cgit_parse_snapshots_mask(value);
         else if (!strcmp(name, "enable-index-links"))
                 ctx.cfg.enable_index_links = atoi(value);
         else if (!strcmp(name, "enable-log-filecount"))
                 ctx.cfg.enable_log_filecount = atoi(value);
         else if (!strcmp(name, "enable-log-linecount"))
                 ctx.cfg.enable_log_linecount = atoi(value);
         else if (!strcmp(name, "max-stats"))
                 ctx.cfg.max_stats = cgit_find_stats_period(value, NULL);
         else if (!strcmp(name, "cache-size"))
                 ctx.cfg.cache_size = atoi(value);
         else if (!strcmp(name, "cache-root"))
                 ctx.cfg.cache_root = xstrdup(value);
         else if (!strcmp(name, "cache-root-ttl"))
                 ctx.cfg.cache_root_ttl = atoi(value);
         else if (!strcmp(name, "cache-repo-ttl"))
                 ctx.cfg.cache_repo_ttl = atoi(value);
         else if (!strcmp(name, "cache-static-ttl"))
                 ctx.cfg.cache_static_ttl = atoi(value);
         else if (!strcmp(name, "cache-dynamic-ttl"))
                 ctx.cfg.cache_dynamic_ttl = atoi(value);
         else if (!strcmp(name, "embedded"))
                 ctx.cfg.embedded = atoi(value);
         else if (!strcmp(name, "max-message-length"))
                 ctx.cfg.max_msg_len = atoi(value);
         else if (!strcmp(name, "max-repodesc-length"))
                 ctx.cfg.max_repodesc_len = atoi(value);
         else if (!strcmp(name, "max-repo-count"))
                 ctx.cfg.max_repo_count = atoi(value);
         else if (!strcmp(name, "max-commit-count"))
                 ctx.cfg.max_commit_count = atoi(value);
         else if (!strcmp(name, "summary-log"))
                 ctx.cfg.summary_log = atoi(value);
         else if (!strcmp(name, "summary-branches"))
                 ctx.cfg.summary_branches = atoi(value);
         else if (!strcmp(name, "summary-tags"))
                 ctx.cfg.summary_tags = atoi(value);
         else if (!strcmp(name, "agefile"))
                 ctx.cfg.agefile = xstrdup(value);
         else if (!strcmp(name, "renamelimit"))
                 ctx.cfg.renamelimit = atoi(value);
         else if (!strcmp(name, "robots"))
                 ctx.cfg.robots = xstrdup(value);
         else if (!strcmp(name, "clone-prefix"))
                 ctx.cfg.clone_prefix = xstrdup(value);
         else if (!strcmp(name, "local-time"))
                 ctx.cfg.local_time = atoi(value);
         else if (!strcmp(name, "repo.group"))
                 ctx.cfg.repo_group = xstrdup(value);
         else if (!strcmp(name, "repo.url"))
                 ctx.repo = cgit_add_repo(value);
         else if (!strcmp(name, "repo.name"))
                 ctx.repo->name = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.path"))
                 ctx.repo->path = trim_end(value, '/');
         else if (ctx.repo && !strcmp(name, "repo.clone-url"))
                 ctx.repo->clone_url = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.desc"))
                 ctx.repo->desc = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.owner"))
                 ctx.repo->owner = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.defbranch"))
                 ctx.repo->defbranch = xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.snapshots"))
                 ctx.repo->snapshots = ctx.cfg.snapshots & cgit_parse_snapshots_mask(value); /* XXX: &? */
         else if (ctx.repo && !strcmp(name, "repo.enable-log-filecount"))
                 ctx.repo->enable_log_filecount = ctx.cfg.enable_log_filecount * atoi(value);
         else if (ctx.repo && !strcmp(name, "repo.enable-log-linecount"))
                 ctx.repo->enable_log_linecount = ctx.cfg.enable_log_linecount * atoi(value);
         else if (ctx.repo && !strcmp(name, "repo.max-stats"))
                 ctx.repo->max_stats = cgit_find_stats_period(value, NULL);
         else if (ctx.repo && !strcmp(name, "repo.module-link"))
                 ctx.repo->module_link= xstrdup(value);
         else if (ctx.repo && !strcmp(name, "repo.readme") && value != NULL) {
                 if (*value == '/')
                         ctx.repo->readme = xstrdup(value);
                 else
                         ctx.repo->readme = xstrdup(fmt("%s/%s", ctx.repo->path, value));
         } else if (!strcmp(name, "include"))
                 parse_configfile(value, config_cb);
 }
 
 static void querystring_cb(const char *name, const char *value)
 {
         if (!strcmp(name,"r")) {
                 ctx.qry.repo = xstrdup(value);
                 ctx.repo = cgit_get_repoinfo(value);
         } else if (!strcmp(name, "p")) {
                 ctx.qry.page = xstrdup(value);
         } else if (!strcmp(name, "url")) {
                 ctx.qry.url = xstrdup(value);
diff --git a/cgit.h b/cgit.h
index 8c64efe..d0fff1f 100644
--- a/cgit.h
+++ b/cgit.h
@@ -4,249 +4,261 @@
 
 #include <git-compat-util.h>
 #include <cache.h>
 #include <grep.h>
 #include <object.h>
 #include <tree.h>
 #include <commit.h>
 #include <tag.h>
 #include <diff.h>
 #include <diffcore.h>
 #include <refs.h>
 #include <revision.h>
 #include <log-tree.h>
 #include <archive.h>
 #include <xdiff-interface.h>
 #include <xdiff/xdiff.h>
 #include <utf8.h>
 
 
 /*
  * Dateformats used on misc. pages
  */
 #define FMT_LONGDATE "%Y-%m-%d %H:%M:%S (%Z)"
 #define FMT_SHORTDATE "%Y-%m-%d"
 #define FMT_ATOMDATE "%Y-%m-%dT%H:%M:%SZ"
 
 
 /*
  * Limits used for relative dates
  */
 #define TM_MIN    60
 #define TM_HOUR  (TM_MIN * 60)
 #define TM_DAY   (TM_HOUR * 24)
 #define TM_WEEK  (TM_DAY * 7)
 #define TM_YEAR  (TM_DAY * 365)
 #define TM_MONTH (TM_YEAR / 12.0)
 
 
 /*
  * Default encoding
  */
 #define PAGE_ENCODING "UTF-8"
 
 typedef void (*configfn)(const char *name, const char *value);
 typedef void (*filepair_fn)(struct diff_filepair *pair);
 typedef void (*linediff_fn)(char *line, int len);
 
 struct cgit_repo {
         char *url;
         char *name;
         char *path;
         char *desc;
         char *owner;
         char *defbranch;
         char *group;
         char *module_link;
         char *readme;
         char *clone_url;
         int snapshots;
         int enable_log_filecount;
         int enable_log_linecount;
         int max_stats;
         time_t mtime;
 };
 
 struct cgit_repolist {
         int length;
         int count;
         struct cgit_repo *repos;
 };
 
 struct commitinfo {
         struct commit *commit;
         char *author;
         char *author_email;
         unsigned long author_date;
         char *committer;
         char *committer_email;
         unsigned long committer_date;
         char *subject;
         char *msg;
         char *msg_encoding;
 };
 
 struct taginfo {
         char *tagger;
         char *tagger_email;
         unsigned long tagger_date;
         char *msg;
 };
 
 struct refinfo {
         const char *refname;
         struct object *object;
         union {
                 struct taginfo *tag;
                 struct commitinfo *commit;
         };
 };
 
 struct reflist {
         struct refinfo **refs;
         int alloc;
         int count;
 };
 
 struct cgit_query {
         int has_symref;
         int has_sha1;
         char *raw;
         char *repo;
         char *page;
         char *search;
         char *grep;
         char *head;
         char *sha1;
         char *sha2;
         char *path;
         char *name;
         char *mimetype;
         char *url;
         char *period;
         int   ofs;
         int nohead;
         char *sort;
         int showmsg;
 };
 
+struct cgit_filter {
+        char *cmd;
+        char **argv;
+        int old_stdout;
+        int pipe_fh[2];
+        int pid;
+        int exitstatus;
+};
+
 struct cgit_config {
         char *agefile;
         char *cache_root;
         char *clone_prefix;
         char *css;
         char *favicon;
         char *footer;
         char *head_include;
         char *header;
         char *index_header;
         char *index_info;
         char *logo;
         char *logo_link;
         char *module_link;
         char *repo_group;
         char *robots;
         char *root_title;
         char *root_desc;
         char *root_readme;
         char *script_name;
         char *virtual_root;
         int cache_size;
         int cache_dynamic_ttl;
         int cache_max_create_time;
         int cache_repo_ttl;
         int cache_root_ttl;
         int cache_static_ttl;
         int embedded;
         int enable_index_links;
         int enable_log_filecount;
         int enable_log_linecount;
         int local_time;
         int max_repo_count;
         int max_commit_count;
         int max_lock_attempts;
         int max_msg_len;
         int max_repodesc_len;
         int max_stats;
         int nocache;
         int noheader;
         int renamelimit;
         int snapshots;
         int summary_branches;
         int summary_log;
         int summary_tags;
 };
 
 struct cgit_page {
         time_t modified;
         time_t expires;
         size_t size;
         char *mimetype;
         char *charset;
         char *filename;
         char *etag;
         char *title;
         int status;
         char *statusmsg;
 };
 
 struct cgit_context {
         struct cgit_query qry;
         struct cgit_config cfg;
         struct cgit_repo *repo;
         struct cgit_page page;
 };
 
 struct cgit_snapshot_format {
         const char *suffix;
         const char *mimetype;
         write_archive_fn_t write_func;
         int bit;
 };
 
 extern const char *cgit_version;
 
 extern struct cgit_repolist cgit_repolist;
 extern struct cgit_context ctx;
 extern const struct cgit_snapshot_format cgit_snapshot_formats[];
 
 extern struct cgit_repo *cgit_add_repo(const char *url);
 extern struct cgit_repo *cgit_get_repoinfo(const char *url);
 extern void cgit_repo_config_cb(const char *name, const char *value);
 
 extern int chk_zero(int result, char *msg);
 extern int chk_positive(int result, char *msg);
 extern int chk_non_negative(int result, char *msg);
 
 extern char *trim_end(const char *str, char c);
 extern char *strlpart(char *txt, int maxlen);
 extern char *strrpart(char *txt, int maxlen);
 
 extern void cgit_add_ref(struct reflist *list, struct refinfo *ref);
 extern int cgit_refs_cb(const char *refname, const unsigned char *sha1,
                         int flags, void *cb_data);
 
 extern void *cgit_free_commitinfo(struct commitinfo *info);
 
 extern int cgit_diff_files(const unsigned char *old_sha1,
                            const unsigned char *new_sha1,
                            unsigned long *old_size, unsigned long *new_size,
                            int *binary, linediff_fn fn);
 
 extern void cgit_diff_tree(const unsigned char *old_sha1,
                            const unsigned char *new_sha1,
                            filepair_fn fn, const char *prefix);
 
 extern void cgit_diff_commit(struct commit *commit, filepair_fn fn);
 
 extern char *fmt(const char *format,...);
 
 extern struct commitinfo *cgit_parse_commit(struct commit *commit);
 extern struct taginfo *cgit_parse_tag(struct tag *tag);
 extern void cgit_parse_url(const char *url);
 
 extern const char *cgit_repobasename(const char *reponame);
 
 extern int cgit_parse_snapshots_mask(const char *str);
 
+extern int cgit_open_filter(struct cgit_filter *filter);
+extern int cgit_close_filter(struct cgit_filter *filter);
+
 
 #endif /* CGIT_H */
diff --git a/shared.c b/shared.c
index cce0af4..288cfa2 100644
--- a/shared.c
+++ b/shared.c
@@ -230,128 +230,163 @@ int filediff_cb(void *priv, mmbuffer_t *mb, int nbuf)
                 if (mb[i].ptr[mb[i].size-1] != '\n') {
                         /* Incomplete line */
                         diffbuf = xrealloc(diffbuf, buflen + mb[i].size);
                         memcpy(diffbuf + buflen, mb[i].ptr, mb[i].size);
                         buflen += mb[i].size;
                         continue;
                 }
 
                 /* we have a complete line */
                 if (!diffbuf) {
                         ((linediff_fn)priv)(mb[i].ptr, mb[i].size);
                         continue;
                 }
                 diffbuf = xrealloc(diffbuf, buflen + mb[i].size);
                 memcpy(diffbuf + buflen, mb[i].ptr, mb[i].size);
                 ((linediff_fn)priv)(diffbuf, buflen + mb[i].size);
                 free(diffbuf);
                 diffbuf = NULL;
                 buflen = 0;
         }
         if (diffbuf) {
                 ((linediff_fn)priv)(diffbuf, buflen);
                 free(diffbuf);
                 diffbuf = NULL;
                 buflen = 0;
         }
         return 0;
 }
 
 int cgit_diff_files(const unsigned char *old_sha1,
                     const unsigned char *new_sha1, unsigned long *old_size,
                     unsigned long *new_size, int *binary, linediff_fn fn)
 {
         mmfile_t file1, file2;
         xpparam_t diff_params;
         xdemitconf_t emit_params;
         xdemitcb_t emit_cb;
 
         if (!load_mmfile(&file1, old_sha1) || !load_mmfile(&file2, new_sha1))
                 return 1;
 
         *old_size = file1.size;
         *new_size = file2.size;
 
         if ((file1.ptr && buffer_is_binary(file1.ptr, file1.size)) ||
             (file2.ptr && buffer_is_binary(file2.ptr, file2.size))) {
                 *binary = 1;
                 return 0;
         }
 
         memset(&diff_params, 0, sizeof(diff_params));
         memset(&emit_params, 0, sizeof(emit_params));
         memset(&emit_cb, 0, sizeof(emit_cb));
         diff_params.flags = XDF_NEED_MINIMAL;
         emit_params.ctxlen = 3;
         emit_params.flags = XDL_EMIT_FUNCNAMES;
         emit_cb.outf = filediff_cb;
         emit_cb.priv = fn;
         xdl_diff(&file1, &file2, &diff_params, &emit_params, &emit_cb);
         return 0;
 }
 
 void cgit_diff_tree(const unsigned char *old_sha1,
                     const unsigned char *new_sha1,
                     filepair_fn fn, const char *prefix)
 {
         struct diff_options opt;
         int ret;
         int prefixlen;
 
         diff_setup(&opt);
         opt.output_format = DIFF_FORMAT_CALLBACK;
         opt.detect_rename = 1;
         opt.rename_limit = ctx.cfg.renamelimit;
         DIFF_OPT_SET(&opt, RECURSIVE);
         opt.format_callback = cgit_diff_tree_cb;
         opt.format_callback_data = fn;
         if (prefix) {
                 opt.nr_paths = 1;
                 opt.paths = &prefix;
                 prefixlen = strlen(prefix);
                 opt.pathlens = &prefixlen;
         }
         diff_setup_done(&opt);
 
         if (old_sha1 && !is_null_sha1(old_sha1))
                 ret = diff_tree_sha1(old_sha1, new_sha1, "", &opt);
         else
                 ret = diff_root_tree_sha1(new_sha1, "", &opt);
         diffcore_std(&opt);
         diff_flush(&opt);
 }
 
 void cgit_diff_commit(struct commit *commit, filepair_fn fn)
 {
         unsigned char *old_sha1 = NULL;
 
         if (commit->parents)
                 old_sha1 = commit->parents->item->object.sha1;
         cgit_diff_tree(old_sha1, commit->object.sha1, fn, NULL);
 }
 
 int cgit_parse_snapshots_mask(const char *str)
 {
         const struct cgit_snapshot_format *f;
         static const char *delim = " \t,:/|;";
         int tl, sl, rv = 0;
 
         /* favor legacy setting */
         if(atoi(str))
                 return 1;
         for(;;) {
                 str += strspn(str,delim);
                 tl = strcspn(str,delim);
                 if (!tl)
                         break;
                 for (f = cgit_snapshot_formats; f->suffix; f++) {
                         sl = strlen(f->suffix);
                         if((tl == sl && !strncmp(f->suffix, str, tl)) ||
                            (tl == sl-1 && !strncmp(f->suffix+1, str, tl-1))) {
                                 rv |= f->bit;
                                 break;
                         }
                 }
                 str += tl;
         }
         return rv;
 }
+
+int cgit_open_filter(struct cgit_filter *filter)
+{
+
+        filter->old_stdout = chk_positive(dup(STDOUT_FILENO),
+                "Unable to duplicate STDOUT");
+        chk_zero(pipe(filter->pipe_fh), "Unable to create pipe to subprocess");
+        filter->pid = chk_non_negative(fork(), "Unable to create subprocess");
+        if (filter->pid == 0) {
+                close(filter->pipe_fh[1]);
+                chk_non_negative(dup2(filter->pipe_fh[0], STDIN_FILENO),
+                        "Unable to use pipe as STDIN");
+                execvp(filter->cmd, filter->argv);
+                die("Unable to exec subprocess %s: %s (%d)", filter->cmd,
+                        strerror(errno), errno);
+        }
+        close(filter->pipe_fh[0]);
+        chk_non_negative(dup2(filter->pipe_fh[1], STDOUT_FILENO),
+                "Unable to use pipe as STDOUT");
+        close(filter->pipe_fh[1]);
+        return 0;
+}
+
+int cgit_close_filter(struct cgit_filter *filter)
+{
+        chk_non_negative(dup2(filter->old_stdout, STDOUT_FILENO),
+                "Unable to restore STDOUT");
+        close(filter->old_stdout);
+        if (filter->pid < 0)
+                return 0;
+        waitpid(filter->pid, &filter->exitstatus, 0);
+        if (WIFEXITED(filter->exitstatus) && !WEXITSTATUS(filter->exitstatus))
+                return 0;
+        die("Subprocess %s exited abnormally", filter->cmd);
+}