Ticket UUID: | 3592747 | |||
Title: | (tailcall|yieldto) failed to find the proper namespace Abort | |||
Type: | Bug | Version: | None | |
Submitter: | coldstore | Created on: | 2012-12-05 08:40:10 | |
Subsystem: | 60. NRE and coroutines | Assigned To: | pooryorick | |
Priority: | 8 | Severity: | Minor | |
Status: | Closed | Last Modified: | 2018-06-28 18:48:32 | |
Resolution: | Fixed | Closed By: | dgp | |
Closed on: | 2018-06-28 18:48:32 | |||
Description: |
Under some circumstances a coroutine can be invoked via ::yieldto while the namespace in which it resides is dying (nsPtr->flags&NS_DYING) This may occur, and is occuring, for example, when a TclOO object is being destroyed, and a coroutine within it is invoked by an event. The call to TclGetNamespaceFromObj() at tclBasic.c:8532 will indicate an error, because the current frame's namespace is not considered to match itself in the case that it is dying. I am unsure what the purpose of the test is, or under what circumstances the two namespaces might differ, but this abort occurs as a result of script evaluation, and hence should be considered a bug. | |||
User Comments: |
dgp added on 2018-06-27 15:48:31:
This is now fixed and the ticket can be closed? pooryorick added on 2018-06-24 07:24:20: (text/x-fossil-wiki) In [9198c16407f3] namespace handling is removed from <code>TclNRTailcallObjCmd</code>, allowing <code>TclNRTailcallObjCmd</code> to handle namespace issues later. This aligns the behaviour of the non-compiled variant with the compiled one: An error is produced if the namespace is not found. pooryorick added on 2018-06-23 21:28:39: (text/x-fossil-wiki) If <code>{*}</code> from the example the compiled path for <code>tailcall</code> is taken, and in the compiled variant an error is raised: <code><verbatim> namespace "::ns" not found </verbatim></code> pooryorick added on 2018-06-23 17:12:25: (text/x-fossil-wiki) A slightly smaller example: <code><verbatim> namespace eval ns { proc p args { ::tailcall [namespace current] {*}$args } namespace delete [namespace current] p } </verbatim></code> pooryorick added on 2018-06-23 17:00:22: (text/x-fossil-wiki) The following script reproduces the issue: <code><verbatim> namespace eval ns { proc p args { ::tailcall [namespace current] {*}$args } namespace eval [namespace current] { namespace delete [namespace current] p } } </verbatim></code> [https://core.tcl.tk/tips/doc/trunk/tip/498.md|TIP 498] is one solution to this issue. dkf added on 2013-03-20 16:27:56: This is why it's great to have a test, so we can confirm when the bug is dead… msofer added on 2013-03-19 21:02:55: Is this bug still present? msofer added on 2012-12-06 09:39:05: The problem is indeed with [tailcall] whose guts are reused by [yieldto]. The current namespace was being cached in an nsNameType Tcl_Obj in order to insure its survival, and as a failure to resolve "could not happen" it was signaled by a panic. As colin is wont to do, he triggered the impossible! Fixed by using a better lifetime management of the namespace. msofer added on 2012-12-05 17:45:00: Note: the problem is likely to appear also with [tailcall]. msofer added on 2012-12-05 17:36:45: 1. no script to repro? 2. any suggestion as to what should happen in that case? Would an error be in order? |