ui-tree: add support for source-filter option

This new option is used to specify an external command which will be
executed when displaying blob content in the tree view. Blob content
will be written to STDIN of the filter and STDOUT from the filter
will be included verbatim in the html output from cgit. The file name
of the blob will be passed as the only argument to the filter command.

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

4 files changed, 25 insertions, 4 deletions

diff --git a/cgit.c b/cgit.c
index 779a464..eb7b45d 100644
--- a/cgit.c
+++ b/cgit.c
@@ -39,128 +39,130 @@ void config_cb(const char *name, const char *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, "source-filter"))
+                ctx.cfg.source_filter = new_filter(value, 1);
         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);
                 cgit_parse_url(value);
         } else if (!strcmp(name, "qt")) {
                 ctx.qry.grep = xstrdup(value);
         } else if (!strcmp(name, "q")) {
diff --git a/cgit.h b/cgit.h
index d0fff1f..f9cf0df 100644
--- a/cgit.h
+++ b/cgit.h
@@ -122,128 +122,129 @@ struct cgit_query {
         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_filter *source_filter;
 };
 
 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);
 
diff --git a/cgitrc.5.txt b/cgitrc.5.txt
index a207fe0..d420ad4 100644
--- a/cgitrc.5.txt
+++ b/cgitrc.5.txt
@@ -137,128 +137,136 @@ max-message-length::
 max-repo-count::
                 Specifies the number of entries to list per page on therepository
         index page. Default value: "50".
 
 max-repodesc-length::
         Specifies the maximum number of repo description characters to display
         on the repository index page. Default value: "80".
 
 max-stats::
         Set the default maximum statistics period. Valid values are "week",
         "month", "quarter" and "year". If unspecified, statistics are
         disabled. Default value: none. See also: "repo.max-stats".
 
 module-link::
         Text which will be used as the formatstring for a hyperlink when a
         submodule is printed in a directory listing. The arguments for the
         formatstring are the path and SHA1 of the submodule commit. Default
         value: "./?repo=%s&page=commit&id=%s"
 
 nocache::
         If set to the value "1" caching will be disabled. This settings is
         deprecated, and will not be honored starting with cgit-1.0. Default
         value: "0".
 
 noheader::
         Flag which, when set to "1", will make cgit omit the standard header
         on all pages. Default value: none. See also: "embedded".
 
 renamelimit::
         Maximum number of files to consider when detecting renames. The value
          "-1" uses the compiletime value in git (for further info, look at
           `man git-diff`). Default value: "-1".
 
 repo.group::
         A value for the current repository group, which all repositories
         specified after this setting will inherit. Default value: none.
 
 robots::
         Text used as content for the "robots" meta-tag. Default value:
         "index, nofollow".
 
 root-desc::
         Text printed below the heading on the repository index page. Default
         value: "a fast webinterface for the git dscm".
 
 root-readme::
         The content of the file specified with this option will be included
         verbatim below the "about" link on the repository index page. Default
         value: none.
 
 root-title::
         Text printed as heading on the repository index page. Default value:
         "Git Repository Browser".
 
 snapshots::
         Text which specifies the default (and allowed) set of snapshot formats
         supported by cgit. The value is a space-separated list of zero or more
         of the following values:
                         "tar"   uncompressed tar-file
                         "tar.gz"gzip-compressed tar-file
                         "tar.bz2"bzip-compressed tar-file
                         "zip"   zip-file
         Default value: none.
 
+source-filter::
+        Specifies a command which will be invoked to format plaintext blobs
+        in the tree view. The command will get the blob content on its STDIN
+        and the name of the blob as its only command line argument. The STDOUT
+        from the command will be included verbatim as the blob contents, i.e.
+        this can be used to implement e.g. syntax highlighting. Default value:
+        none.
+
 summary-branches::
         Specifies the number of branches to display in the repository "summary"
         view. Default value: "10".
 
 summary-log::
         Specifies the number of log entries to display in the repository
         "summary" view. Default value: "10".
 
 summary-tags::
         Specifies the number of tags to display in the repository "summary"
         view. Default value: "10".
 
 virtual-root::
         Url which, if specified, will be used as root for all cgit links. It
         will also cause cgit to generate 'virtual urls', i.e. urls like
         '/cgit/tree/README' as opposed to '?r=cgit&p=tree&path=README'. Default
         value: none.
         NOTE: cgit has recently learned how to use PATH_INFO to achieve the
         same kind of virtual urls, so this option will probably be deprecated.
 
 REPOSITORY SETTINGS
 -------------------
 repo.clone-url::
         A list of space-separated urls which can be used to clone this repo.
         Default value: none.
 
 repo.defbranch::
         The name of the default branch for this repository. If no such branch
         exists in the repository, the first branch name (when sorted) is used
         as default instead. Default value: "master".
 
 repo.desc::
         The value to show as repository description. Default value: none.
 
 repo.enable-log-filecount::
         A flag which can be used to disable the global setting
         `enable-log-filecount'. Default value: none.
 
 repo.enable-log-linecount::
         A flag which can be used to disable the global setting
         `enable-log-linecount'. Default value: none.
 
 repo.max-stats::
         Override the default maximum statistics period. Valid values are equal
         to the values specified for the global "max-stats" setting. Default
         value: none.
 
 repo.name::
         The value to show as repository name. Default value: <repo.url>.
 
 repo.owner::
         A value used to identify the owner of the repository. Default value:
         none.
 
 repo.path::
         An absolute path to the repository directory. For non-bare repositories
         this is the .git-directory. Default value: none.
 
 repo.readme::
         A path (relative to <repo.path>) which specifies a file to include
         verbatim as the "About" page for this repo. Default value: none.
 
 repo.snapshots::
         A mask of allowed snapshot-formats for this repo, restricted by the
diff --git a/ui-tree.c b/ui-tree.c
index 553dbaa..816e121 100644
--- a/ui-tree.c
+++ b/ui-tree.c
@@ -1,262 +1,272 @@
 /* ui-tree.c: functions for tree output
  *
  * Copyright (C) 2006 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
 #include <ctype.h>
 #include "cgit.h"
 #include "html.h"
 #include "ui-shared.h"
 
 char *curr_rev;
 char *match_path;
 int header = 0;
 
-static void print_text_buffer(char *buf, unsigned long size)
+static void print_text_buffer(const char *name, char *buf, unsigned long size)
 {
         unsigned long lineno, idx;
         const char *numberfmt =
                 "<a class='no' id='n%1$d' name='n%1$d' href='#n%1$d'>%1$d</a>\n";
 
         html("<table summary='blob content' class='blob'>\n");
+        if (ctx.cfg.source_filter) {
+                html("<tr><td class='lines'><pre><code>");
+                ctx.cfg.source_filter->argv[1] = xstrdup(name);
+                cgit_open_filter(ctx.cfg.source_filter);
+                write(STDOUT_FILENO, buf, size);
+                cgit_close_filter(ctx.cfg.source_filter);
+                html("</code></pre></td></tr></table>\n");
+                return;
+        }
+
         html("<tr><td class='linenumbers'><pre>");
         idx = 0;
         lineno = 0;
 
         if (size) {
                 htmlf(numberfmt, ++lineno);
                 while(idx < size - 1) { // skip absolute last newline
                         if (buf[idx] == '\n')
                                 htmlf(numberfmt, ++lineno);
                         idx++;
                 }
         }
         html("</pre></td>\n");
         html("<td class='lines'><pre><code>");
         html_txt(buf);
         html("</code></pre></td></tr></table>\n");
 }
 
 #define ROWLEN 32
 
 static void print_binary_buffer(char *buf, unsigned long size)
 {
         unsigned long ofs, idx;
         static char ascii[ROWLEN + 1];
 
         html("<table summary='blob content' class='bin-blob'>\n");
         html("<tr><th>ofs</th><th>hex dump</th><th>ascii</th></tr>");
         for (ofs = 0; ofs < size; ofs += ROWLEN, buf += ROWLEN) {
                 htmlf("<tr><td class='right'>%04x</td><td class='hex'>", ofs);
                 for (idx = 0; idx < ROWLEN && ofs + idx < size; idx++)
                         htmlf("%*s%02x",
                               idx == 16 ? 4 : 1, "",
                               buf[idx] & 0xff);
                 html(" </td><td class='hex'>");
                 for (idx = 0; idx < ROWLEN && ofs + idx < size; idx++)
                         ascii[idx] = isgraph(buf[idx]) ? buf[idx] : '.';
                 ascii[idx] = '\0';
                 html_txt(ascii);
                 html("</td></tr>\n");
         }
         html("</table>\n");
 }
 
-static void print_object(const unsigned char *sha1, char *path)
+static void print_object(const unsigned char *sha1, char *path, const char *basename)
 {
         enum object_type type;
         char *buf;
         unsigned long size;
 
         type = sha1_object_info(sha1, &size);
         if (type == OBJ_BAD) {
                 cgit_print_error(fmt("Bad object name: %s",
                                      sha1_to_hex(sha1)));
                 return;
         }
 
         buf = read_sha1_file(sha1, &type, &size);
         if (!buf) {
                 cgit_print_error(fmt("Error reading object %s",
                                      sha1_to_hex(sha1)));
                 return;
         }
 
         html(" (");
         cgit_plain_link("plain", NULL, NULL, ctx.qry.head,
                         curr_rev, path);
         htmlf(")<br/>blob: %s\n", sha1_to_hex(sha1));
 
         if (buffer_is_binary(buf, size))
                 print_binary_buffer(buf, size);
         else
-                print_text_buffer(buf, size);
+                print_text_buffer(basename, buf, size);
 }
 
 
 static int ls_item(const unsigned char *sha1, const char *base, int baselen,
                    const char *pathname, unsigned int mode, int stage,
                    void *cbdata)
 {
         char *name;
         char *fullpath;
         enum object_type type;
         unsigned long size = 0;
 
         name = xstrdup(pathname);
         fullpath = fmt("%s%s%s", ctx.qry.path ? ctx.qry.path : "",
                        ctx.qry.path ? "/" : "", name);
 
         if (!S_ISGITLINK(mode)) {
                 type = sha1_object_info(sha1, &size);
                 if (type == OBJ_BAD) {
                         htmlf("<tr><td colspan='3'>Bad object: %s %s</td></tr>",
                               name,
                               sha1_to_hex(sha1));
                         return 0;
                 }
         }
 
         html("<tr><td class='ls-mode'>");
         cgit_print_filemode(mode);
         html("</td><td>");
         if (S_ISGITLINK(mode)) {
                 htmlf("<a class='ls-mod' href='");
                 html_attr(fmt(ctx.repo->module_link,
                               name,
                               sha1_to_hex(sha1)));
                 html("'>");
                 html_txt(name);
                 html("</a>");
         } else if (S_ISDIR(mode)) {
                 cgit_tree_link(name, NULL, "ls-dir", ctx.qry.head,
                                curr_rev, fullpath);
         } else {
                 cgit_tree_link(name, NULL, "ls-blob", ctx.qry.head,
                                curr_rev, fullpath);
         }
         htmlf("</td><td class='ls-size'>%li</td>", size);
 
         html("<td>");
         cgit_log_link("log", NULL, "button", ctx.qry.head, curr_rev,
                       fullpath, 0, NULL, NULL, ctx.qry.showmsg);
         if (ctx.repo->max_stats)
                 cgit_stats_link("stats", NULL, "button", ctx.qry.head,
                                 fullpath);
         html("</td></tr>\n");
         free(name);
         return 0;
 }
 
 static void ls_head()
 {
         html("<table summary='tree listing' class='list'>\n");
         html("<tr class='nohover'>");
         html("<th class='left'>Mode</th>");
         html("<th class='left'>Name</th>");
         html("<th class='right'>Size</th>");
         html("<th/>");
         html("</tr>\n");
         header = 1;
 }
 
 static void ls_tail()
 {
         if (!header)
                 return;
         html("</table>\n");
         header = 0;
 }
 
 static void ls_tree(const unsigned char *sha1, char *path)
 {
         struct tree *tree;
 
         tree = parse_tree_indirect(sha1);
         if (!tree) {
                 cgit_print_error(fmt("Not a tree object: %s",
                                      sha1_to_hex(sha1)));
                 return;
         }
 
         ls_head();
         read_tree_recursive(tree, "", 0, 1, NULL, ls_item, NULL);
         ls_tail();
 }
 
 
 static int walk_tree(const unsigned char *sha1, const char *base, int baselen,
                      const char *pathname, unsigned mode, int stage,
                      void *cbdata)
 {
         static int state;
         static char buffer[PATH_MAX];
         char *url;
 
         if (state == 0) {
                 memcpy(buffer, base, baselen);
                 strcpy(buffer+baselen, pathname);
                 url = cgit_pageurl(ctx.qry.repo, "tree",
                                    fmt("h=%s&amp;path=%s", curr_rev, buffer));
                 html("/");
                 cgit_tree_link(xstrdup(pathname), NULL, NULL, ctx.qry.head,
                                curr_rev, buffer);
 
                 if (strcmp(match_path, buffer))
                         return READ_TREE_RECURSIVE;
 
                 if (S_ISDIR(mode)) {
                         state = 1;
                         ls_head();
                         return READ_TREE_RECURSIVE;
                 } else {
-                        print_object(sha1, buffer);
+                        print_object(sha1, buffer, pathname);
                         return 0;
                 }
         }
         ls_item(sha1, base, baselen, pathname, mode, stage, NULL);
         return 0;
 }
 
 
 /*
  * Show a tree or a blob
  *   rev:  the commit pointing at the root tree object
  *   path: path to tree or blob
  */
 void cgit_print_tree(const char *rev, char *path)
 {
         unsigned char sha1[20];
         struct commit *commit;
         const char *paths[] = {path, NULL};
 
         if (!rev)
                 rev = ctx.qry.head;
 
         curr_rev = xstrdup(rev);
         if (get_sha1(rev, sha1)) {
                 cgit_print_error(fmt("Invalid revision name: %s", rev));
                 return;
         }
         commit = lookup_commit_reference(sha1);
         if (!commit || parse_commit(commit)) {
                 cgit_print_error(fmt("Invalid commit reference: %s", rev));
                 return;
         }
 
         html("path: <a href='");
         html_attr(cgit_pageurl(ctx.qry.repo, "tree", fmt("h=%s", rev)));
         html("'>root</a>");
 
         if (path == NULL) {
                 ls_tree(commit->tree->object.sha1, NULL);
                 return;
         }
 
         match_path = path;
         read_tree_recursive(commit->tree, NULL, 0, 0, paths, walk_tree, NULL);
         ls_tail();
 }