Review main Merged

25814[BUGFIX] ClassAliasMap, Tx_ VH namespace and closing tag throws Exception

Claus Due CR +2 V +2 Easy +1 −0 updated 12/04/13, 18:54

Commit message

When using aliased ViewHelper class names and old Tx_ namespace in
template and ViewHelper uses closing tag (not self-closing) an
error is thrown, saying closing tag ViewHelper is not the same
as openening tag ViewHelper (closing tag uses old class name,
opening tag uses new). To solve, TemplateParser now checks if
resolved ViewHelper class names are aliases of other classes and
if so, uses the real class name instead of the alias.

Steps to reproduce error:

* template namespace: {namespace myext=Tx_Myext_ViewHelpers}
* template code: <myext:vh>test</myext:vh>
* namespaced VH class: \Myext\ViewHelpers\VhViewHelper
* ClassAliasMap: Tx_Myext_ViewHelpers_VhViewHelper ->
  \Mext\ViewHelpers\VhViewHelper
* framework: render template using any View

Error 1224485398 "closing tag does not match opening tag" thrown.

Steps taken to fix error:

* run constructed class name through alias resolve method.

Review

  • Code review CR +2
    • Helmut Hummel +2
  • Verified V +2
    • Helmut Hummel +2

Files 1

Review log 28

Show 3 earlier messages
  1. Mr. Jenkins commented on patch set 1 11/30/13, 21:31

  2. Claus Due uploaded patch set 2 11/30/13, 22:52

  3. Claus Due commented on patch set 2 11/30/13, 22:53

    Patch set 2: instead of early returning $realClassName, $name is now set to the value before being returned - to be more consistent with other code in the same class.
  1. Mr. Jenkins commented on patch set 2 11/30/13, 23:05

  2. Claus Due uploaded patch set 3 12/01/13, 13:09

  3. Claus Due commented on patch set 3 12/01/13, 13:10

    Updated commit message with clear description to reproduce and steps taken to solve.
  4. Mr. Jenkins commented on patch set 3 12/01/13, 13:23

  5. Claus Due uploaded patch set 4 12/02/13, 21:45

  6. Claus Due uploaded patch set 5 12/02/13, 21:46

  7. Claus Due uploaded patch set 6 12/02/13, 21:50

  8. Claus Due rebased onto patch set 6 12/02/13, 21:58

  9. Mr. Jenkins commented on patch set 4 12/02/13, 21:58

  10. Mr. Jenkins commented on patch set 5 12/02/13, 21:59

  11. Mr. Jenkins commented on patch set 6 12/02/13, 22:12

  12. Mr. Jenkins commented on patch set 7 12/02/13, 22:14

  13. Ludwig Rafelsberger commented on patch set 7 CR +1 12/02/13, 22:33

    Reconstructed the problem as described. The patch clearly solves that error for me. I didn't further test the system: lacking experience and infrastructure ;-(…
  14. Georg Ringer voted on patch set 7 V +1 CR +1 12/03/13, 05:06

  15. Wouter Wolters voted on patch set 7 CR +1 12/03/13, 18:31

  16. Danilo Bürger commented on patch set 7 V +1 CR +1 12/03/13, 21:23

    Reproduced on 6.1.5 / 6.1.6, commit fixes it
  17. Helmut Hummel commented on patch set 7 CR +1 12/04/13, 17:50

    could be merged in master now if super important, but no backports until merge freeze is over
    should also go into 6.0
  18. Claus Due commented on patch set 7 12/04/13, 17:52

    @Helmut I would -greatly- appreciate merging to master. Our CI setup and further namespace migrations in fx EXT:flux depend on it. We can certainly live without the patch on 6.1 for a while still :)
  19. Helmut Hummel commented on patch set 7 12/04/13, 18:24

    @Claus: you take care of the backports (also 6.0) once the merge freeze is over and I will merge it in master. Deal?
  20. Claus Due commented on patch set 7 12/04/13, 18:41

    @Helmut Deal - I've made a calendar entry with alert so I remember it. And thanks! :)
  21. Claus Due commented on patch set 7 12/04/13, 18:42

    Ah, also 6.0 - will update commit with Releases:
  22. Claus Due uploaded patch set 8 12/04/13, 18:42

  23. Mr. Jenkins commented on patch set 8 12/04/13, 18:53

  24. Helmut Hummel commented on patch set 8 V +2 CR +2 12/04/13, 18:54

    last patch set only updated commit message. votes for ps before count
  25. Helmut Hummel merged patch set 9 12/04/13, 18:54