summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
| | * | | No longer document (even though it's only in an example) that ↵jan.nijtmans2017-11-162-3/+3
| | |\ \ \ | | | | |/ | | | |/| | | | | | Tcl_SavedResult is a struct, and that the internal representation of an int is stored in the object's internalRep.longValue member. That might no longer be true in the future.
| | | * | No longer document (even though it's only in an example) that ↵jan.nijtmans2017-11-162-9/+8
| | | | | | | | | | | | | | | | | | | | Tcl_SavedResult is a struct, and that the internal representation of an int is stored in the object's internalRep.longValue member. That might no longer be true in the future.
| * | | | Remove compat/float.h and related machinery. The last system known where ↵jan.nijtmans2017-11-1510-43/+6
| | | | | | | | | | | | | | | | | | | | | | | | | this was needed was SunOS-4, which is not supported by Tcl any more for a long ... long time .... Also, fix a typo in generic/tclInt.h and remove some end-of-line spacing.
| * | | | Change signature of (internal) TclScanElement() function. This saves memory ↵jan.nijtmans2017-11-155-32/+15
| |\ \ \ \ | | |/ / / | | | | | | | | | | allocation and the possibility for panic's in dict and list handling, requiring 1/4 of memory for internal allocation of temporary storage. No change to external API.
| | * | | Change signature of (internal) TclScanElement() function. This saves memory ↵jan.nijtmans2017-11-155-32/+15
| | | | | | | | | | | | | | | | | | | | allocation and the possibility for panic's in dict and list handling, requiring 1/4 of memory for internal allocation of temporary storage. No change to external API.
| * | | | Repair lost of broken [package prefer] testing.dgp2017-11-141-25/+100
| | | | |
| * | | | [5d6de65036] [package require] with [package prefer stable] was not choosing ↵dgp2017-11-142-46/+81
| |\ \ \ \ | | |/ / / | | | | | | | | | | available stable package.
| | * | | [5d6de65036] [package require] with [package prefer stable] was not choosing ↵dgp2017-11-142-35/+68
| | |\ \ \ | | | |/ / | | | | | | | | | | available stable package.
| | | * | [5d6de65036] [package require] with [package prefer stable] was not choosingdgp2017-11-142-33/+68
| | | |\ \ | | | | | | | | | | | | available stable package.
| | | | * | Ticket [5d65e65036]. My fix. Do not skip the second check for stable ↵aku2017-11-131-33/+58
| | | | | | | | | | | | | | | | | | | | | | | | versions even when the main check fails. Avoided radical changes to the structure (kept single search loop, with selection after).
| | | | * | Test case for Bug 5d65e65036.dgp2017-11-061-0/+10
| | | | | |
| * | | | | Merge core-8-6-branchjan.nijtmans2017-11-133-7/+1
| |\ \ \ \ \ | | |/ / / /
| | * | | | No longer mark Tcl_EvalFile() as obsolete, since it will continue to be ↵jan.nijtmans2017-11-132-2/+0
| | |\ \ \ \ | | | |/ / / | | | | | | | | | | | | supported in Tcl 9.0
| | | * | | No longer mark Tcl_EvalFile() as obsolete, since it will continue to be ↵jan.nijtmans2017-11-132-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | supported in Tcl 9.0
| * | | | | Add a lot of "deprecation" marks to internal API which will be removed in ↵jan.nijtmans2017-11-134-66/+63
| | | | | | | | | | | | | | | | | | | | | | | | Tcl 9.0 (see also tcl-9-cleanup branch). Extensions still using this API will get a compiler warning, but everything will continue to work fine for all future 8.x releases.
| * | | | | merge core-8-6-branchjan.nijtmans2017-11-102-3/+23
| |\ \ \ \ \ | | |/ / / /
| | * | | | Make "string split" and "string is (alpha|graph|...)" work as expected with ↵jan.nijtmans2017-11-102-3/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Unicode chars > U+ffff, when Tcl is compiled with TCL_UTF_MAX == 4. No effect when TCL_UTF_MAX == 3 or TCL_UTF_MAX == 6. Test-case added for "string split".
| * | | | | merge core-8-6-branchjan.nijtmans2017-11-092-5/+13
| |\ \ \ \ \ | | |/ / / /
| | * | | | Make "scan %c" and the internal function ExtendUnicodeRepWithString() work ↵jan.nijtmans2017-11-092-5/+13
| | | | | | | | | | | | | | | | | | | | | | | | as expected for TCL_UTF_MAX == 4 when handling characters > U+ffff. No effect when TCL_UTF_MAX == 3 or TCL_UTF_MAX == 6
| * | | | | For alpha/beta versions of Tcl, stub-enabled extensions are only allowed to ↵jan.nijtmans2017-11-091-5/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | run on the EXACT the same Tcl versions as compiled: The stubs might still be in flux, they are not guaranteed to work on future Tcl versions anyway. Backported from "novem" (which already had this special protection). For final releases (8.7.0) everything works as before.
| * | | | | Change "epoch" field for dicts from "int" to "unsigned int". This doubles ↵jan.nijtmans2017-11-082-11/+11
| | | | | | | | | | | | | | | | | | | | | | | | the number of available epoch's before overflow. Also make 0 the special value in stead of -1, meaning "dict search done". That also means that epoch counting starts with 1.
| * | | | | compiler warningdgp2017-11-081-1/+1
| |\ \ \ \ \ | | |/ / / /
| | * | | | compiler warningdgp2017-11-081-1/+1
| | |\ \ \ \ | | | |/ / /
| | | * | | compiler warningdgp2017-11-081-1/+1
| | | | | |
| * | | | | merge 8.6dgp2017-11-082-6/+12
| |\ \ \ \ \ | | |/ / / /
| | * | | | [3298012] Stop RebuildTable asking ckalloc for more than it can give.dgp2017-11-081-5/+8
| | |\ \ \ \ | | | |/ / /
| | | * | | [3298012] Stop RebuildTable asking ckalloc for more than it can give.dgp2017-11-081-5/+8
| | | |\ \ \ | | | | |/ / | | | |/| |
| | | | * | Attempted bug fix.dgp2017-11-071-5/+8
| | | |/ /
| | * | | TclOO object allocation: Set classPtr to NULL if it wasn't otherwise set.dgp2017-11-081-1/+4
| | | | |
| * | | | Modify TclCreateProc to handle arbitrary argument names, not just ASCII.dgp2017-11-083-57/+45
| | | | |
| * | | | merge core-8-6-branchjan.nijtmans2017-11-072-15/+15
| |\ \ \ \ | | |/ / / | | | | / | | |_|/ | |/| |
| | * | Don't use Unicode character \udead in test-cases, because it is an invalid ↵jan.nijtmans2017-11-072-15/+15
| | | | | | | | | | | | | | | | Unicode code-point (lower surrogate). This will cause test-failures with TIP #389. Just use \ud0ad in stead. Also fix some other test-cases which fail for TCL_UTF_MAX == 4.
| * | | Fix version number of Windows Help file (not sure if it's actually used, but ↵jan.nijtmans2017-11-061-2/+2
| | | | | | | | | | | | | | | | better safe than sorry)
| * | | update .project file with branch name. Make clear that optparse doesnt work ↵jan.nijtmans2017-11-053-3/+3
| | | | | | | | | | | | | | | | with 8.4 any more
| * | | [0d902eadd6] Ensure that [string first] handles ASCII in Unicode correctly.dkf2017-11-043-23/+40
| |\ \ \
| | * | | Fix for the weird [string first] behaviour.dkf2017-11-041-19/+23
| | | | |
| | * | | Detected bug in [string first] with unicode. Pat Thoyts found it.dkf2017-11-042-4/+17
| |/ / /
| * | | Revise a few stray packages not yet ready to tolerate a Tcl 9 bump.dgp2017-11-035-9/+9
| | | |
| * | | TIP 345 Implementation.dgp2017-11-033-39/+78
| |\ \ \
| | * \ \ merge trunkdgp2017-10-3010-37/+65
| | |\ \ \
| | * \ \ \ merge trunkdgp2017-10-2728-395/+425
| | |\ \ \ \
| | * | | | | Create new testing command [teststringbytes] to probe the things thatdgp2017-10-262-11/+47
| | | | | | | | | | | | | | | | | | | | | otherwise require [encoding convertto identity]. adapt encoding-15.3 to use.
| | * | | | | Convert remaining tests to use [testbytestring].dgp2017-10-251-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | encoding-15.3 still needs replacement for [encoding convertto identity]. That is, some testing command to expose objPtr->bytes.
| | * | | | | Repair Tcl_CreateEncoding(); Modernize [testencoding];dgp2017-10-253-13/+12
| | | | | | | | | | | | | | | | | | | | | Update most tests toying with identity encoding.
| | * | | | | Stop using "identity" as an encoding to test basic functionng of thedgp2017-10-241-4/+4
| | | | | | | | | | | | | | | | | | | | | [encoding] command. "iso8859-1" is another one always available.
| | * | | | | Extend Tcl_CreateEncoding() to be able to create an encoding withoutdgp2017-10-241-17/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | registering it when it has no name. Then use this to create "identity" encoding without naming it "identity". Then no one can look it up by that name.
| | * | | | | backout initial commit; need more care.dgp2017-10-231-0/+77
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Binary writes internally make use of this encoding. Need to hide it instead of destroy it.
| | * | | | | Implementation branch for TIP 345: Kill the "identity" encoding.dgp2017-10-231-77/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This checkin, completely does that. Bad news is it causes huge failures. We're still making a lot of use of this encoding, and cannot complete this TIP until the branch is repaired.
| * | | | | | merge core-8-6-branchjan.nijtmans2017-11-032-3/+8
| |\ \ \ \ \ \ | | | |_|_|/ / | | |/| | | |
| | * | | | | Fix [6f2f83cc149e9918884faffefebc8dfa695f4ea0|6f2f83cc14]: tclWinload.c ↵jan.nijtmans2017-11-032-3/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | robustness. And fix a minor possible memory leak in TclSetupEnv() as well. Thanks to Christian Werner for both suggestions, backported from Androwish.
generated by cgit v0.12 at 2024-11-16 21:56:49 (GMT)