asdf merge requestshttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests2023-06-05T14:17:56Zhttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/216Use :darwin feature versus :target-os-darwin2023-06-05T14:17:56ZTarn Burtontwburton@gmail.comUse :darwin feature versus :target-os-darwinClasp supports the features specified by trivial-features automatically. We are planning on removing `:target-os-darwin`, `:target-os-linux`, etc.Clasp supports the features specified by trivial-features automatically. We are planning on removing `:target-os-darwin`, `:target-os-linux`, etc.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/200Add flexi-streams to *known-systems-with-bad-secondary-system-names*2022-01-10T04:05:27ZEric TimmonsAdd flexi-streams to *known-systems-with-bad-secondary-system-names*See https://github.com/edicl/flexi-streams/issues/25See https://github.com/edicl/flexi-streams/issues/25https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/186abcl: fix loading asdf systems from jar files2021-08-24T05:29:45ZMark Evensonabcl: fix loading asdf systems from jar filesUnder ABCL, when loading system definitions from a JAR-PATHNAME, the
PARSE-UNIX-NAMESTRING was including the DEVICE in the defaults to
MAKE-PATHNAME. This was mostly harmless but fails when (the as yet
unlrelease abcl-1.8.1) tightened th...Under ABCL, when loading system definitions from a JAR-PATHNAME, the
PARSE-UNIX-NAMESTRING was including the DEVICE in the defaults to
MAKE-PATHNAME. This was mostly harmless but fails when (the as yet
unlrelease abcl-1.8.1) tightened the definition of JAR-PATHNAME to
always have an absolute directoryhttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/185WIP: Makefile target for the Debian package.2021-07-29T00:16:37ZRobert GoldmanWIP: Makefile target for the Debian package.Reverse-engineered from tools/release.lisp. **This is incomplete** -- I couldn't figure out the check for a clean build directory, partly because `git` does not set the exit status.
I don't have access to a Debian machine to test the b...Reverse-engineered from tools/release.lisp. **This is incomplete** -- I couldn't figure out the check for a clean build directory, partly because `git` does not set the exit status.
I don't have access to a Debian machine to test the build on, even once that's fixed, I'm afraid. @etimmons could you have a shot?https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/183Do not trigger the standard pipeline and New image pipeline simultaneously2022-04-27T22:19:37ZEric TimmonsDo not trigger the standard pipeline and New image pipeline simultaneouslySee for example: https://gitlab.common-lisp.net/asdf/asdf/-/pipelines/4202
I didn't catch this before because we tested the New image pipeline triggers on
another branch.See for example: https://gitlab.common-lisp.net/asdf/asdf/-/pipelines/4202
I didn't catch this before because we tested the New image pipeline triggers on
another branch.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/177Rework package-local-nicknames so that uiop/package package is not modified2021-06-04T02:54:39ZEric TimmonsRework package-local-nicknames so that uiop/package package is not modifieduiop/package now defines functions in its own package corresponding to the PLN
API (that simply call the implementation's native functions). Then, once
UIOP:DEFINE-PACKAGE is established, a new package is defined that simply
reexports th...uiop/package now defines functions in its own package corresponding to the PLN
API (that simply call the implementation's native functions). Then, once
UIOP:DEFINE-PACKAGE is established, a new package is defined that simply
reexports the implementation's native PLN API.
Fixes #723.3.5https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/162Draft: Resolve "The dictatorship of versioning"2021-04-18T21:17:33ZEric TimmonsDraft: Resolve "The dictatorship of versioning"Closes #50
When compared to the approach I described in #50, this doesn't have `component-raw-version`. I realized it's unnecessary as we're already exporting the `version` symbol, so extensions can access that slot directly.
This MR ...Closes #50
When compared to the approach I described in #50, this doesn't have `component-raw-version`. I realized it's unnecessary as we're already exporting the `version` symbol, so extensions can access that slot directly.
This MR no longer restricts the contents of the `version` slot. Existing functions for reading the slot (`component-version` and `system-version`) now only return the contents of the slot if it meets ASDF preexisting requirements (string of dot-separated natural numbers).
Make version normalization a GF so it can be extended. Add `system-aesthetic-version-string` to allow extensions to provide a string representation of their version objects.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/159On Genera running under macOS, do not search /usr/share/common-lisp for ASDF ...2021-02-16T16:47:49ZGary PalterOn Genera running under macOS, do not search /usr/share/common-lisp for ASDF systemsmacOS has implemented a separate read-only system volume and /usr/share is on that volume.
Consequently, /usr/share/common-lisp does not exist nor can it be created. Further, Genera
accesses its host filesystem through NFS and there is n...macOS has implemented a separate read-only system volume and /usr/share is on that volume.
Consequently, /usr/share/common-lisp does not exist nor can it be created. Further, Genera
accesses its host filesystem through NFS and there is no way to export the root (/) or
/usr/share that will not lead Genera into an infinite loop trying to access those mounts.
Instead, just search /usr/local/share which is on the writable Data volume.
Also, change the value return by UIOP/OS:ARCHITECTURE to :VLM when running on a VLM.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/158On Genera running under macOS, do not search /usr/share/common-lisp for ASDF ...2021-02-15T20:28:08ZGary PalterOn Genera running under macOS, do not search /usr/share/common-lisp for ASDF systemsmacOS has implemented a separate read-only system volume and /usr/share is on that volume.
Consequently, /usr/share/common-lisp does not exist nor can it be created. Further, Genera
accesses its host filesystem through NFS and there is n...macOS has implemented a separate read-only system volume and /usr/share is on that volume.
Consequently, /usr/share/common-lisp does not exist nor can it be created. Further, Genera
accesses its host filesystem through NFS and there is no way to export the root (/) or
/usr/share that will not lead Genera into an infinite loop trying to access those mounts.
Instead, just search /usr/local/share which is on the writable Data volume.
Also, change the value return by UIOP/OS:ARCHITECTURE to :VLM when running on a VLM.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/148Set the correct slot name error-output-stream, not error-stream.2023-12-11T18:39:54ZMartin SimmonsSet the correct slot name error-output-stream, not error-stream.`launch-program` contains three incorrect uses of `error-stream`, which should be `error-output-stream`.`launch-program` contains three incorrect uses of `error-stream`, which should be `error-output-stream`.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/144LAUNCH-PROGRAM: fix ERROR-OUTPUT-STREAM slot name2020-11-19T10:36:30ZLuís OliveiraLAUNCH-PROGRAM: fix ERROR-OUTPUT-STREAM slot namePROCESS-INFO contains an ERROR-OUTPUT-STREAM rather than an
ERROR-STREAM slot.PROCESS-INFO contains an ERROR-OUTPUT-STREAM rather than an
ERROR-STREAM slot.Robert GoldmanRobert Goldmanhttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/141Improve interaction between find-definition and WITH-UPGRADABILITY2020-05-18T06:29:58ZLuís OliveiraImprove interaction between find-definition and WITH-UPGRADABILITYIf we don't discard the original defun or defgeneric form from the
expansion, find-definition is able to pin point the correct form
instead of falling back to the parent WITH-UPGRADABILITY form.If we don't discard the original defun or defgeneric form from the
expansion, find-definition is able to pin point the correct form
instead of falling back to the parent WITH-UPGRADABILITY form.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/137Ignore errors in sb-grovel-unknown-constant-condition-p2020-04-30T18:43:12ZEric TimmonsIgnore errors in sb-grovel-unknown-constant-condition-pAn error is raised if `simple-condition-format-control` is a formatter function
and is is passed to `string`. This now mirrors `match-condition-p` which also
ignores errors when comparing a simple condition to a string.An error is raised if `simple-condition-format-control` is a formatter function
and is is passed to `string`. This now mirrors `match-condition-p` which also
ignores errors when comparing a simple condition to a string.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/134bundle: use coerce-filename in output-files2020-02-02T19:39:26ZFrançois-René Rideaubundle: use coerce-filename in output-filesAddresses #25.Addresses #25.Robert GoldmanRobert Goldmanhttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/128Allow directories to be treated as fasls for clasp2019-12-19T04:44:35ZChristian SchafmeisterAllow directories to be treated as fasls for claspThis merge request is to allow directories to be treated as
fasl files in Clasp Common Lisp.
Clasp Common Lisp uses C dynamic libraries as fasl files.
On macOS, DWARF debug info is separated from dynamic libraries
into special DWARF fil...This merge request is to allow directories to be treated as
fasl files in Clasp Common Lisp.
Clasp Common Lisp uses C dynamic libraries as fasl files.
On macOS, DWARF debug info is separated from dynamic libraries
into special DWARF files that sit in the filesystem next to
the dynamic libraries. These DWARF files are created by passing
the dynamic library to the 'dsymutil' utility.
This causes problems with atomic creation of fasls.
So in Clasp we want to use directories as fasl files.
/foo/bar.fasl will become a directory and on macOS will contain
/foo/bar.fasl/fasl.fasl and /foo/bar.fasl/fasl.fasl.DWARF.
But ASDF is currently hardcoded to only allow fasls to be files.
So this merge request relaxes the requirement that fasls be files in
ASDF but only for clasp by adding four #-clasp feature tests.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/125Fixes several errors in compile-file* for clasp2020-01-15T17:35:27ZKarsten PoeckFixes several errors in compile-file* for clasp* drop useless rename that renamed a file to the same name
* compile first to a *.o based on the temp-name
* rename *.o, *.bc, *.dwarf in addition to *.fasl
Did run make test l=clasp, gives the same 7 errors as before this change* drop useless rename that renamed a file to the same name
* compile first to a *.o based on the temp-name
* rename *.o, *.bc, *.dwarf in addition to *.fasl
Did run make test l=clasp, gives the same 7 errors as before this changehttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/124Add a TEST-OP-TEST-FAILURE condition for test libraries to sub-class2020-01-15T17:34:31ZVladimir Sedachvsedach@common-lisp.netAdd a TEST-OP-TEST-FAILURE condition for test libraries to sub-classA condition class that test libraries can signal so that ASDF:TEST-SYSTEM can report test failures.
See the mailing list discussion:
https://mailman.common-lisp.net/pipermail/asdf-devel/2019-September/thread.htmlA condition class that test libraries can signal so that ASDF:TEST-SYSTEM can report test failures.
See the mailing list discussion:
https://mailman.common-lisp.net/pipermail/asdf-devel/2019-September/thread.htmlhttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/115Master2018-12-18T06:26:48ZPascal J. BourguignonMasterhttps://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/113Add subsection for the :entry-point option2018-10-19T17:54:24ZHugo IshimaruAdd subsection for the :entry-point option1. Since the BNF of defsystem grammar lacked the `:entry-point` option, I added a simple subsection for it. And then I put some `@anchor`s in this section for convenience.
2. I reclassified two error classes to the type index.1. Since the BNF of defsystem grammar lacked the `:entry-point` option, I added a simple subsection for it. And then I put some `@anchor`s in this section for convenience.
2. I reclassified two error classes to the type index.https://gitlab.common-lisp.net/asdf/asdf/-/merge_requests/105Support for system virtual slots.2018-10-19T14:56:58ZDidier VernaSupport for system virtual slots.Here's a implementation of system virtual slots, as discussed in MR103.
A system virtual slot is a slot the value of which is read either from the
system directly, or potentially from its primary system if needed. It avoids
duplicating ...Here's a implementation of system virtual slots, as discussed in MR103.
A system virtual slot is a slot the value of which is read either from the
system directly, or potentially from its primary system if needed. It avoids
duplicating meta-information (license, author etc.) in secondary systems.
As a side-effect of this implementation, the functions
system{-long}-description now behave virtually, as opposed to
component{-long}-description. Also, a new function called system-version is
provided, which also behaves virtually, as opposed to component-version.
2018-09-18 Didier Verna <didier@didierverna.net>
* interface.lisp (:asdf/interface): Export SYSTEM-VERSION.
* system.lisp (:asdf/system): Ditto.
(system): Only automatically define generic writers for virtual
slots.
(*system-virtual-slots*): New variable. List the virtual slot names.
(system-virtual-slot-value): New function. General virtual slot
reader.
(define-system-virtual-slot-reader): New macro. Define specific
virtual slot readers based on the above function.
(define-system-virtual-slot-readers): New macro. Define all
specific virtual slot readers.
(system-license): New ad-hoc virtual slot reader, defined manually
in addition to system-licenCe (note the 'c').3.3.3Robert GoldmanRobert Goldman