- up

[packages/rpm-build-tools.git] / pldnotify.awk

diff --git a/pldnotify.awk b/pldnotify.awk
index 9a01bcb9465525eef11e04091a7554a4eeb2deb7..0cf980233c4df21f0defa84dc4f8f814f495813b 100644 (file)
--- a/pldnotify.awk
+++ b/pldnotify.awk
@@ -1,4 +1,4 @@
-#!/usr/bin/gawk -f
+#!/bin/awk -f
 # $Revision$, $Date$
 #
 # Copyright (C) 2000-2010 PLD-Team <feedback@pld-linux.org>
@@ -22,8 +22,8 @@ function d(s) {
                return
        }
 
-       print strftime("%Y-%m-%d %H:%M:%S ") s >> "/dev/stderr"
-#      print s >> "/dev/stderr"
+#      print strftime("%Y-%m-%d %H:%M:%S ") s >> "/dev/stderr"
+       print s >> "/dev/stderr"
 }
 
 function fixedsub(s1,s2,t,     ind) {
@@ -193,6 +193,26 @@ function postfix_link(url, link,   oldlink) {
        return link
 }
 
+# use perl HTML::TreeBuilder module to extract links from html
+# it returns TAGNAME LINK in output which is pretty stright forward to parse in awk
+function extract_links_cmd(tmpfile) {
+       return "perl -MHTML::TreeBuilder -e ' \
+       my $content = join q//, <>; \
+       my $root = new HTML::TreeBuilder; \
+       $root->parse($content); \
+       \
+       my %links = (); \
+       for (@{$root->extract_links(qw(a iframe))}) { \
+               my($link, $element, $attr, $tag) = @$_; \
+               $links{$link} = $tag; \
+       } \
+       \
+       while (my($link, $tag) = each %links) { \
+               print $tag, q/ /, $link, $/; \
+       } \
+       ' " tmpfile
+}
+
 # get all <A HREF=..> tags from specified URL
 function get_links(url,filename,   errno,link,oneline,retval,odp,wholeodp,lowerodp,tmpfile,cmd) {
 
@@ -201,11 +221,9 @@ function get_links(url,filename,   errno,link,oneline,retval,odp,wholeodp,lowero
        tmpfile = mktemp()
        tmpfileerr = mktemp()
 
-       if (url ~ /^http:\/\/(download|downloads|dl)\.(sf|sourceforge)\.net\//) {
+       if (url ~ /^http:\/\/(download|dl)\.(sf|sourceforge)\.net\//) {
                # http://downloads.sourceforge.net/project/mediainfo/source/mediainfo/
                gsub("^http://(download|dl)\.(sf|sourceforge)\.net/", "", url)
-               # http://downloads.sourceforge.net/project/mediainfo/source/mediainfo/
-               gsub("^http://downloads\.(sf|sourceforge)\.net/project/", "", url)
 
                gsub("/.*", "", url)
                url = "http://sourceforge.net/projects/" url "/files/"
@@ -272,6 +290,43 @@ function get_links(url,filename,   errno,link,oneline,retval,odp,wholeodp,lowero
                retval = ("WGET ERROR: " errno ": " wholeerr)
                return retval
        }
+       system("rm -f " tmpfileerr)
+
+       urldir = url;
+       sub(/[^\/]+$/, "", urldir)
+
+if (USE_PERL) {
+       cmd = extract_links_cmd(tmpfile)
+       while (cmd | getline) {
+               tag = $1
+               link = substr($0, length(tag) + 2)
+
+               if (tag == "iframe") {
+                       d("Frame: " link)
+                       if (url !~ /\//) {
+                               link = (urldir link)
+                               d("Frame->: " link)
+                       }
+
+                       if (link_seen(link)) {
+                               continue
+                       }
+                       retval = (retval " " get_links(link))
+               }
+
+               if (link_seen(link)) {
+                       continue
+               }
+
+               retval = (retval " " link)
+               d("href(): " link)
+       }
+       close(cmd)
+       system("rm -f " tmpfile)
+
+       d("Returning: [" retval "]")
+       return retval
+}
 
        wholeodp = ""
        d("Reading success response...")
@@ -280,12 +335,7 @@ function get_links(url,filename,   errno,link,oneline,retval,odp,wholeodp,lowero
 #              d("Response: " wholeodp)
        }
        d("Reponse read done...")
-
        system("rm -f " tmpfile)
-       system("rm -f " tmpfileerr)
-
-       urldir = url;
-       sub(/[^\/]+$/, "", urldir)
 
        while (match(wholeodp, /<([aA]|[fF][rR][aA][mM][eE])[ \t][^>]*>/) > 0) {
                d("Processing links...")
@@ -359,7 +409,7 @@ function get_links(url,filename,   errno,link,oneline,retval,odp,wholeodp,lowero
                }
        }
 
-       d("Returning: " retval)
+       d("Returning: [" retval "]")
        return retval
 }
 
@@ -404,12 +454,12 @@ function find_mirror(url) {
        return url
 }
 
-function process_source(number,lurl,name,version) {
+function process_source(number, lurl, name, version) {
 # fetches file list, and compares version numbers
        d("Processing " lurl)
 
-       if ( index(lurl,version)==0 ) {
-               d("There is no version number.")
+       if (index(lurl, version) == 0) {
+               d("There is no version number ["version"] in ["lurl"]")
                return 0
        }
 
@@ -494,7 +544,7 @@ function process_source(number,lurl,name,version) {
                                }
                        }
                }
-               if (finished==0)
+               if (finished == 0)
                        print name "(" number ") seems ok: " oldversion
                else
                        print name "(" number ") [OLD] " oldversion " [NEW] " version
@@ -565,7 +615,7 @@ function vim_upgrade(name, ver,     mver, nver, vimcmd) {
        }
 }
 
-function process_data(name,ver,rel,src) {
+function process_data(name, ver, rel, src) {
        if (name ~ /^php-pear-/) {
                return pear_upgrade(name, ver);
        }
@@ -578,16 +628,20 @@ function process_data(name,ver,rel,src) {
        if (name == "vim") {
                return vim_upgrade(name, ver);
        }
+       if (name == "xulrunner") {
+               ver = subst_defines(DEFS["firefox_ver"], DEFS)
+               d("package xulrunner, change version to firefox ["ver"]")
+       }
 
 # this function checks if substitutions were valid, and if true:
 # processes each URL and tries to get current file list
        for (i in src) {
-               if ( src[i] ~ /%{nil}/ ) {
+               if (src[i] ~ /%{nil}/) {
                        gsub(/\%\{nil\}/, "", src[i])
                }
                if ( src[i] !~ /%{.*}/ && src[i] !~ /%[A-Za-z0-9_]/ )  {
                        d("Source: " src[i])
-                       process_source(i,src[i],name,ver)
+                       process_source(i, src[i], name, ver)
                } else {
                        print FNAME ":" i ": impossible substitution: " src[i]
                }