I have improved the test case now. I wasn't aware you can copy it back
from the rejected queue (or forgot), and I irregularly delete past
uploads, so that could have meant recreating it.

Most of the time, if a test plan is lacking a bit, the SRU gets accepted
with a note to add some missing bits to the test plan.


** Description changed:

  [Impact]
  The packaging does not respect compiler flags, notably hardening flags like 
the stack protector, as well as frame pointer, as it never uses dpkg-buildflags 
and is still old manual style.
  
  This means there is a higher security risk, which given the nature of
  the library seems ill-advised, and the lack of frame pointers hampers
  profiling.
  
  [Test plan]
  Look at the build log and see that the build flags have been passed, notably 
stuff like
  
- -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O2 -Wall -g -O2 
-Werror=implicit-function-declaration -fno-omit-frame-pointer 
-mno-omit-leaf-frame-pointer -ffile-prefix-map=/<<PKGBUILDDIR>>=. 
-fstack-protector-strong -fstack-clash-protection -Wformat -Werror=for
- mat-security -fcf-protection
+ -fno-strict-overflow
+ -Werror=implicit-function-declaration
+ -fno-omit-frame-pointer
+ -mno-omit-leaf-frame-pointer
+ -fstack-protector-strong
+ -fstack-clash-protection 
+ -Wformat 
+ -Werror=for
  
- Parse some xml and see that it works
+ Run https://gist.github.com/julian-
+ klode/85e55553f85c410a1b856a93dce77208 on the ELF binaries and ensure
+ that all lines have -fno-omit-frame-pointer
+ 
+ 
+ Run a smoke test to see the parser still works:
+ 
+ from lxml import etree
+ from urllib.request import urlopen
+ 
+ with urlopen("https://www.w3.org/TR/REC-xml/REC-xml-20081126.xml";) as xml:
+     xml=xml.read()
+ tree=etree.fromstring(xml, 
base_url="https://www.w3.org/TR/REC-xml/REC-xml-20081126.xml";)
+ 
+ for element in tree.iter():
+     print(element)
+ 
+ print(etree.tostring(tree))
+ 
  
  [Where problems could occur]
  Hardening flags could surface existing (possibly security) bugs that have 
been asymptomatic so far; frame pointers will incur some slow down.

** Description changed:

  [Impact]
  The packaging does not respect compiler flags, notably hardening flags like 
the stack protector, as well as frame pointer, as it never uses dpkg-buildflags 
and is still old manual style.
  
  This means there is a higher security risk, which given the nature of
  the library seems ill-advised, and the lack of frame pointers hampers
  profiling.
  
  [Test plan]
- Look at the build log and see that the build flags have been passed, notably 
stuff like
+ Skim at the build log and see that the build flags have been passed, notably 
stuff like
  
  -fno-strict-overflow
  -Werror=implicit-function-declaration
  -fno-omit-frame-pointer
  -mno-omit-leaf-frame-pointer
  -fstack-protector-strong
- -fstack-clash-protection 
- -Wformat 
+ -fstack-clash-protection
+ -Wformat
  -Werror=for
  
  Run https://gist.github.com/julian-
  klode/85e55553f85c410a1b856a93dce77208 on the ELF binaries and ensure
  that all lines have -fno-omit-frame-pointer
- 
  
  Run a smoke test to see the parser still works:
  
  from lxml import etree
  from urllib.request import urlopen
  
  with urlopen("https://www.w3.org/TR/REC-xml/REC-xml-20081126.xml";) as xml:
-     xml=xml.read()
+     xml=xml.read()
  tree=etree.fromstring(xml, 
base_url="https://www.w3.org/TR/REC-xml/REC-xml-20081126.xml";)
  
  for element in tree.iter():
-     print(element)
+     print(element)
  
  print(etree.tostring(tree))
  
- 
  [Where problems could occur]
  Hardening flags could surface existing (possibly security) bugs that have 
been asymptomatic so far; frame pointers will incur some slow down.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2076419

Title:
  lxml: Does not respect compiler flags

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/lxml/+bug/2076419/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to