Skip to contentSkip to conversation

Error handling fixes

PATCH New Paul Eggleton 1 participant 3 messages May 29, 2014

lore
Share & tools
lore

Paul Eggleton

author
#0 lore raw

cover Error handling fixes

The following changes since commit 0e7b594ccbceb3149f38776cea204807031ef69f:

fetch2/hg: Fix missing proto param for hg checkout with user and pw (2014-05-28 08:28:17 +0100)

are available in the git repository at:

git://git.yoctoproject.org/poky-contrib paule/bb-errors
http://git.yoctoproject.org/cgit.cgi/poky-contrib/log/?h=paule/bb-errors

Paul Eggleton (2):
fetch2: improve handling of two classes of URL parameter mistakes
utils: avoid printing traceback on ExpansionError during parsing

 lib/bb/fetch2/__init__.py | 14 ++++++++++----
 lib/bb/utils.py           |  2 ++
 2 files changed, 12 insertions(+), 4 deletions(-)
Signature · 1 line
1.9.3
author
#1 lore raw

PATCH 1/2 fetch2: improve handling of two classes of URL parameter mistakes

Handle the following situations in a URL (e.g. in SRC_URI):

* Trailing semicolon in a URL - this is now ignored.
* Parameter specified with no value (no equals sign). This still produces an error, but at least it is MalformedUrl with a proper message rather than "ValueError: need more than 1 value to unpack".

Signed-off-by Paul Eggleton <[email protected]>
---
 lib/bb/fetch2/__init__.py | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
lib/bb/fetch2/__init__.py +10 -4
index f571fc4..dcada12 100644
@@ -56,8 +56,11 @@ class BBFetchException(Exception):
 
 class MalformedUrl(BBFetchException):
     """Exception raised when encountering an invalid url"""
-    def __init__(self, url):
-         msg = "The URL: '%s' is invalid and cannot be interpreted" % url
+    def __init__(self, url, message=''):
+         if message:
+             msg = message
+         else:
+             msg = "The URL: '%s' is invalid and cannot be interpreted" % url
          self.url = url
          BBFetchException.__init__(self, msg)
          self.args = (url,)

@@ -371,8 +374,11 @@ def decodeurl(url):
     p = {}
     if parm:
         for s in parm.split(';'):
-            s1, s2 = s.split('=')
-            p[s1] = s2
+            if s:
+                if not '=' in s:
+                    raise MalformedUrl(url, "The URL: '%s' is invalid: parameter %s does not specify a value (missing '=')" % (url, s))
+                s1, s2 = s.split('=')
+                p[s1] = s2
 
     return type, host, urllib.unquote(path), user, pswd, p
Signature · 1 line
1.9.3
author
#2 lore raw

PATCH 2/2 utils: avoid printing traceback on ExpansionError during parsing

In reply to Paul Eggleton

If an ExpansionError occurs during better_exec() we should just raise it instead of printing the traceback, so that recipe errors (such as broken URLs in SRC_URI) are more easily comprehensible.

Signed-off-by Paul Eggleton <[email protected]>
---
 lib/bb/utils.py | 2 ++
 1 file changed, 2 insertions(+)
lib/bb/utils.py +2 -0
index 1be1874..ead5f36 100644
@@ -357,6 +357,8 @@ def better_exec(code, context, text = None, realfile = "<code>"):
     except bb.BBHandledException:
         # Error already shown so passthrough
         raise
+    except bb.data_smart.ExpansionError:
+        raise
     except Exception as e:
         (t, value, tb) = sys.exc_info()
Signature · 1 line
1.9.3