bitbake: fetch/wget: Add latest_versionstring method

Being able to query whether updated versions of a url are available
is useful, not least for the package reporting system. Since such code
is closely linked to the url type and the url itself, the fetcher
makes a locical place to contain this code.

For wget based urls this means taking upstream directory listings
and searching those for later versions, returning those that are
found.

The patch also adds unittests for this function so that if
improvements are made, the original test urls can be used
to evaulate the those changes.

This is based on code from Irina Patru <irina.patru@intel.com>.

(Bitbake rev: a8272e22b7819e0e8afd8e291d276f5f28fc0007)

Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

2 files changed, 274 insertions, 0 deletions

diff --git a/bitbake/lib/bb/fetch2/wget.py b/bitbake/lib/bb/fetch2/wget.py
index 0456490368..db5f27b6f1 100644
--- a/bitbake/lib/bb/fetch2/wget.py
+++ b/bitbake/lib/bb/fetch2/wget.py
@@ -25,6 +25,9 @@ BitBake build tools.
 #
 # Based on functions from the base bb module, Copyright 2003 Holger Schurig
 
+import re
+import tempfile
+import subprocess
 import os
 import logging
 import bb
@@ -34,6 +37,7 @@ from   bb.fetch2 import FetchMethod
 from   bb.fetch2 import FetchError
 from   bb.fetch2 import logger
 from   bb.fetch2 import runfetchcmd
+from   bs4 import BeautifulSoup
 
 class Wget(FetchMethod):
     """Class to fetch urls via 'wget'"""
@@ -104,3 +108,238 @@ class Wget(FetchMethod):
         self._runwget(ud, d, fetchcmd, True)
 
         return True
+
+
+    def _parse_path(self, regex, s):
+        """
+        Find and group name, version and archive type in the given string s
+        """
+        bb.debug(3, "parse_path(%s, %s)" % (regex.pattern, s))
+        m = regex.search(s)
+        if m:
+            bb.debug(3, "%s, %s, %s" % (m.group('name'), m.group('ver'), m.group('type')))
+            return (m.group('name'), m.group('ver'), m.group('type'))
+        return None
+
+    def _modelate_version(self, version):
+        if version[0] in ['.', '-']:
+            if version[1].isdigit():
+                version = version[1] + version[0] + version[2:len(version)]
+            else:
+                version = version[1:len(version)]
+
+        version = re.sub('\-', '.', version)
+        version = re.sub('_', '.', version)
+        version = re.sub('(rc)+', '.-1.', version)
+        version = re.sub('(alpha)+', '.-3.', version)
+        version = re.sub('(beta)+', '.-2.', version)
+        if version[0] == 'v':
+            version = version[1:len(version)]
+        return version
+
+    def _vercmp(self, old, new):
+        """
+        Check whether 'new' is newer than 'old' version. We use existing vercmp() for the
+        purpose. PE is cleared in comparison as it's not for build, and PR is cleared too
+        for simplicity as it's somehow difficult to get from various upstream format
+        """
+
+        (oldpn, oldpv, oldsuffix) = old
+        (newpn, newpv, newsuffix) = new
+
+        """
+        Check for a new suffix type that we have never heard of before
+        """
+        if (newsuffix):
+            m = self.suffixregex.search(newsuffix)
+            if not m:
+                bb.warn("%s has a possible unknown suffix: %s" % (newpn, newsuffix))
+                return False
+
+        """
+        Not our package so ignore it
+        """
+        if oldpn != newpn:
+            return False
+
+        oldpv = self._modelate_version(oldpv)
+        newpv = self._modelate_version(newpv)
+
+        if bb.utils.vercmp(("0", oldpv, ""), ("0", newpv, "")) < 0:
+            return True
+        else:
+            return False
+
+    def _fetch_index(self, uri, ud, d):
+        """
+        Run fetch checkstatus to get directory information
+        """
+        f = tempfile.NamedTemporaryFile(dir="/tmp/s/", delete=False)
+
+        agent = "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.12) Gecko/20101027 Ubuntu/9.10 (karmic) Firefox/3.6.12"
+        fetchcmd = self.basecmd
+        fetchcmd += " -O " + f.name + " --user-agent='" + agent + "' '" + uri + "'"
+        try:
+            self._runwget(ud, d, fetchcmd, True)
+            fetchresult = f.read()
+        except bb.fetch2.BBFetchException:
+            fetchresult = ""
+
+        f.close()
+        # os.unlink(f.name)
+        return fetchresult
+
+    def _check_latest_dir(self, url, versionstring, ud, d):
+        """
+        Return the name of the directory with the greatest package version
+        If error or no version, return ""
+        """
+        bb.debug(3, "DirURL: %s, %s" % (url, versionstring))
+        soup = BeautifulSoup(self._fetch_index(url, ud, d))
+        if not soup:
+            return ""
+
+        valid = 0
+        prefix = ''
+        regex = re.compile("(\D*)((\d+[\.\-_])+(\d+))")
+        m = regex.search(versionstring)
+        if m:
+            version = ('', m.group(2), '')
+            prefix = m.group(1)
+            bb.debug(3, "version: %s, prefix: %s" % (version, prefix))
+        else:
+            version = ('', versionstring, '')
+
+        for href in soup.find_all('a', href=True):
+            bb.debug(3, "href: %s" % (href['href']))
+            if href['href'].find(versionstring) >= 0:
+                valid = 1
+            m = regex.search(href['href'].strip("/"))
+            if m:
+                thisversion = ('', m.group(2), '')
+                if thisversion and self._vercmp(version, thisversion) == True:
+                    version = thisversion
+
+        if valid:
+            bb.debug(3, "Would return %s" % (prefix+version[1]))
+            return prefix+version[1]
+        else:
+            bb.debug(3, "Not Valid")
+            return ""
+
+    def _check_latest_version(self, url, packagename, ud, d):
+        """
+        Return the latest version of a package inside a given directory path
+        If error or no version, return ""
+        """
+        valid = 0
+        version = self._parse_path(self.name_version_type_regex, packagename)
+
+        bb.debug(3, "VersionURL: %s" % (url))
+        soup = BeautifulSoup(self._fetch_index(url, ud, d))
+        if not soup:
+            bb.debug(3, "*** %s NO SOUP" % (packagename))
+            return ""
+
+        pn_regex = d.getVar('REGEX', True)
+        if pn_regex:
+            testversion = version
+            pn_regex = re.compile(pn_regex)
+            bb.debug(3, "pn_regex = '%s'" % (pn_regex.pattern))
+            
+        for line in soup.find_all('a', href=True):
+            newver = ('', '', '')
+            bb.debug(3, "line = '%s'" % (line['href']))
+            if pn_regex:
+                m = pn_regex.search(line['href'])
+                if m:
+                    bb.debug(3, "Name = '%s', Pver = '%s'" % (m.group('name'), m.group('pver')))
+                    newver = (m.group('name'), m.group('pver'), '')
+                else:
+                    continue
+            else:
+                newver = self._parse_path(self.name_version_type_regex, line['href'])
+            valid = 1
+            if newver and self._vercmp(version, newver) == True:
+                version = newver
+                
+        # check whether a valid package and version were found
+
+        if not valid:
+            version = ('', '', '')
+        if not pn_regex:
+            testversion = ('', '', '')
+        bb.debug(3, "*** %s -> %s (TestVersion = %s)" % (packagename, version[1], testversion[1]))
+        if valid and version:
+            return re.sub('_', '.', version[1])
+
+    def latest_versionstring(self, ud, d):
+        """
+        Manipulate the URL and try to obtain the latest package version
+
+        sanity check to ensure same name and type. Match as many patterns as possible
+        such as:
+                gnome-common-2.20.0.tar.gz (most common format)
+                gtk+-2.90.1.tar.gz
+                xf86-input-synaptics-12.6.9.tar.gz
+                dri2proto-2.3.tar.gz
+                blktool_4.orig.tar.gz
+                libid3tag-0.15.1b.tar.gz
+                unzip552.tar.gz
+                icu4c-3_6-src.tgz
+                genext2fs_1.3.orig.tar.gz
+                gst-fluendo-mp3
+        """
+        # match most patterns which uses "-" as separator to version digits
+        pn_prefix1 = "[a-zA-Z][a-zA-Z0-9]*([\-_][a-zA-Z]\w+)*\+?[\-_]"
+        # a loose pattern such as for unzip552.tar.gz
+        pn_prefix2 = "[a-zA-Z]+"
+        # a loose pattern such as for 80325-quicky-0.4.tar.gz
+        pn_prefix3 = "[0-9]+[\-]?[a-zA-Z]+"
+        # Save the Package Name (pn) Regex for use later
+        self.pn_regex = "(%s|%s|%s)" % (pn_prefix1, pn_prefix2, pn_prefix3)
+
+        # match version
+        version_regex = "(([A-Z]*\d+[a-zA-Z]*[\.\-_]*)+)"
+
+        # src.rpm extension was added only for rpm package. Can be removed if the rpm
+        # packaged will always be considered as having to be manually upgraded
+        suffixlist = "(tar\.gz|tgz|tar\.bz2|zip|xz|rpm|bz2|orig\.tar\.gz|tar\.xz|src\.tar\.gz|src\.tgz|svnr\d+\.tar\.bz2|stable\.tar\.gz|src\.rpm)"
+        self.suffixregex = re.compile(suffixlist)
+
+        # match name, version and archive type of a package
+        self.name_version_type_regex = re.compile("(?P<name>%s?)\.?v?(?P<ver>%s)(\-source)?[\.\-](?P<type>%s$)" % (self.pn_regex, version_regex, suffixlist))
+
+        regex_uri = d.getVar("REGEX_URI", True)
+        newpath = ud.path
+        pupver = ""
+
+        # search for version matches on folders inside the path, like:
+        # "5.7" in http://download.gnome.org/sources/${PN}/5.7/${PN}-${PV}.tar.gz
+        m = re.search("(?P<dirver>[^/]*(\d+\.)*\d+([\-_]r\d+)*)/", ud.path)
+        bb.debug(3, "path = %s" % (ud.path))
+        bb.debug(3, "Regex: %s" % (self.name_version_type_regex.pattern))
+        if m and not regex_uri:
+            dirver = m.group('dirver')
+            # generate the new uri after removing version directory name
+            newuri = bb.fetch.encodeurl([ud.type, ud.host, ud.path.split(dirver)[0], ud.user, ud.pswd, {}])
+            newversion = self._check_latest_dir(newuri, dirver, ud, d)
+            if newversion and dirver != newversion:
+                newpath = ud.path.replace(dirver, newversion, True)
+
+        # try to acquire all remote files in current directory
+        packagename = newpath.split("/")[-1]            # current package name
+        newpath = newpath.split(packagename)[0] or "/"  # path to directory
+
+        # generate the new uri with the appropriate latest directory
+        newuri = regex_uri or bb.fetch.encodeurl([ud.type, ud.host, newpath, ud.user, ud.pswd, {}])
+        newversion = self._check_latest_version(newuri, packagename, ud, d)
+        while not newversion:
+            # maybe it's hiding in a download directory so try there
+            newuri = "/".join(newuri.split("/")[0:-2]) + "/download"
+            if newuri == "/download" or newuri == "http://download":
+                break
+            newversion = self._check_latest_version(newuri, packagename, ud, d)
+
+        return newversion
+
diff --git a/bitbake/lib/bb/tests/fetch.py b/bitbake/lib/bb/tests/fetch.py
index aa9b222612..9d45743fdf 100644
--- a/bitbake/lib/bb/tests/fetch.py
+++ b/bitbake/lib/bb/tests/fetch.py
@@ -600,6 +600,31 @@ class FetchMethodTest(FetcherTest):
             : "3.82+dbg0.9",
     }
 
+    test_wget_uris = {
+        # packages with versions inside directory name
+        ("util-linux", "http://kernel.org/pub/linux/utils/util-linux/v2.23/util-linux-2.24.2.tar.bz2", "", "")
+            : "2.24.2",
+        ("enchant", "http://www.abisource.com/downloads/enchant/1.6.0/enchant-1.6.0.tar.gz", "", "")
+            : "1.6.0",
+        ("cmake", "http://www.cmake.org/files/v2.8/cmake-2.8.12.1.tar.gz", "", "")
+            : "2.8.12.1",
+        # packages with versions only in current directory
+        ("eglic", "http://downloads.yoctoproject.org/releases/eglibc/eglibc-2.18-svnr23787.tar.bz2", "", "")
+            : "2.19",
+        ("gnu-config", "http://downloads.yoctoproject.org/releases/gnu-config/gnu-config-20120814.tar.bz2", "", "")
+            : "20120814",
+        # packages with "99" in the name of possible version
+        ("pulseaudio", "http://freedesktop.org/software/pulseaudio/releases/pulseaudio-4.0.tar.xz", "", "")
+            : "5.0",
+        ("xserver-xorg", "http://xorg.freedesktop.org/releases/individual/xserver/xorg-server-1.15.1.tar.bz2", "", "")
+            : "1.15.1",
+        # packages with valid REGEX_URI and REGEX
+        ("cups", "http://www.cups.org/software/1.7.2/cups-1.7.2-source.tar.bz2", "http://www.cups.org/software.php", "/software\.php\?VERSION=2\.0\.0&FILE=2\.0\.0/(?P<name>cups\-)(?P<pver>((\d+[\.\-_]*)+))\-source\.tar\.gz")
+            : "2.0.0",
+        ("db", "http://download.oracle.com/berkeley-db/db-5.3.21.tar.gz", "http://www.oracle.com/technetwork/products/berkeleydb/downloads/index-082944.html", "http://download.oracle.com/otn/berkeley-db/(?P<name>db-)(?P<pver>((\d+[\.\-_]*)+))\.tar\.gz")
+            : "6.1.19",
+    }
+
     def test_git_latest_versionstring(self):
         for k, v in self.test_git_uris.items():
             self.d.setVar("SRCREV", k[2])
@@ -609,3 +634,13 @@ class FetchMethodTest(FetcherTest):
             print("Package %s, version: %s <= %s" % (k[0], v, verstring))
             r = bb.utils.vercmp_string(v, verstring)
             self.assertTrue(r == -1 or r == 0)
+
+    def test_wget_latest_versionstring(self):
+        for k, v in self.test_wget_uris.items():
+            self.d.setVar("REGEX_URI", k[2])
+            self.d.setVar("REGEX", k[3])
+            ud = bb.fetch2.FetchData(k[1], self.d)
+            verstring = ud.method.latest_versionstring(ud, self.d)
+            print("Package %s, version: %s <= %s" % (k[0], v, verstring))
+            r = bb.utils.vercmp_string(v, verstring)
+            self.assertTrue(r == -1 or r == 0)