[PATCH] XmlParser: allow use HTTPS URL as recipe file

[Jiri Prochazka]

Current lxml.etree.parse implementation allows using HTTP, but not
HTTPS. When user tries to use HTTPS URL, it crashes on IOException.

This patch fixes that by checking if the URL starts with HTTPS
and uses urlopen to open the HTTPS page. This allows lxml.etree.parse
method to parse the XML from the HTTPS URL.

urllib2 is already used Common/Path and is available since Python2.6
so no new dependency is added.

Signed-off-by: Jiri Prochazka 
---
 lnst/Controller/XmlParser.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/lnst/Controller/XmlParser.py b/lnst/Controller/XmlParser.py
index afc2cda..355b5e8 100644
--- a/lnst/Controller/XmlParser.py
+++ b/lnst/Controller/XmlParser.py
@@ -15,6 +15,7 @@ import re
 import sys
 import copy
 from lxml import etree
+from urllib2 import urlopen
 from lnst.Common.Config import lnst_config
 from lnst.Controller.XmlTemplates import XmlTemplates
 from lnst.Controller.XmlProcessing import XmlProcessingError
@@ -79,7 +80,10 @@ class XmlParser(object):

def _parse(self, path):
         try:
-            doc = etree.parse(path)
+            if path.startswith('https'):
+                doc = etree.parse(urlopen(path))
+            else:
+                doc = etree.parse(path)
         except etree.LxmlError as err:
             # A workaround for cases when lxml (quite strangely)
             # sets the filename to <string>.

[Jan Tluka]

Fri, Mar 11, 2016 at 11:24:22AM CET, [e-mail] wrote:
> Current lxml.etree.parse implementation allows using HTTP, but not
> HTTPS. When user tries to use HTTPS URL, it crashes on IOException.
>
> This patch fixes that by checking if the URL starts with HTTPS
> and uses urlopen to open the HTTPS page. This allows lxml.etree.parse
> method to parse the XML from the HTTPS URL.
>
> urllib2 is already used Common/Path and is available since Python2.6
> so no new dependency is added.
> Signed-off-by: Jiri Prochazka
>
> lnst/Controller/XmlParser.py | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/lnst/Controller/XmlParser.py b/lnst/Controller/XmlParser.py
> index afc2cda..355b5e8 100644
> --- a/lnst/Controller/XmlParser.py
> +++ b/lnst/Controller/XmlParser.py
> @@ -15,6 +15,7 @@ import re
> import sys
> import copy
> from lxml import etree
> +from urllib2 import urlopen
> from lnst.Common.Config import lnst_config
> from lnst.Controller.XmlTemplates import XmlTemplates
> from lnst.Controller.XmlProcessing import XmlProcessingError
> @@ -79,7 +80,10 @@ class XmlParser(object):
> def _parse(self, path):
>     try:
>        doc = etree.parse(path)
>        if path.startswith('https'):
>            doc = etree.parse(urlopen(path))
>        else:
>            doc = etree.parse(path)
>   except etree.LxmlError as err:
>       # A workaround for cases when lxml (quite strangely)
>       # sets the filename to <string>.
>
> -- 
> 2.4.3
> _______________________________________________
> LNST-developers mailing list
> [e-mail]
> https://lists.fedorahosted.org/admin/lists/[e-mail]...

Acked-by: Jan Tluka

[Ondrej Lichtner]

On Fri, Mar 11, 2016 at 11:24:22AM +0100, Jiri Prochazka wrote:
> Current lxml.etree.parse implementation allows using HTTP, but not
> HTTPS. When user tries to use HTTPS URL, it crashes on IOException.
>
> This patch fixes that by checking if the URL starts with HTTPS
> and uses urlopen to open the HTTPS page. This allows lxml.etree.parse
> method to parse the XML from the HTTPS URL.
>
> urllib2 is already used Common/Path and is available since Python2.6
> so no new dependency is added.
>
> Signed-off-by: Jiri Prochazka

Acked-by: Ondrej Lichtner

[Jan Tluka]

Fri, Mar 11, 2016 at 11:24:22AM CET, [e-mail] wrote:
> Current lxml.etree.parse implementation allows using HTTP, but not
> HTTPS. When user tries to use HTTPS URL, it crashes on IOException.
>
> This patch fixes that by checking if the URL starts with HTTPS
> and uses urlopen to open the HTTPS page. This allows lxml.etree.parse
> method to parse the XML from the HTTPS URL.
>
> urllib2 is already used Common/Path and is available since Python2.6
> so no new dependency is added.
> Signed-off-by: Jiri Prochazka

Applied. Thanks!

-Jan