Add accessors for st_blocks and st_blksize #78

DanielG · 2016-11-02T02:24:56Z

AFAIU st_blocks is mandatory in POSIX but st_blksize is part of the XSI option (whatever that means) so it's definition is protected by a feature test macro.

This depends on https://ghc.haskell.org/trac/ghc/ticket/12795.

hvr · 2016-11-02T22:30:33Z

Fyi, I need to think about the conditional API part; we're trying hard to avoid conditional exports going forward with unix, c.f. #23

DanielG · 2016-12-12T20:05:05Z

How about this then?

hvr · 2016-12-12T20:18:12Z

@DanielG better... otoh, if you know it's conditional, you can also make it into a Maybe, that way it can be caught w/o exception handling

hvr · 2016-12-12T20:19:25Z

System/Posix/Files/Common.hsc

+  unsafePerformIO $ withForeignPtr stat $ (#peek struct stat, st_blksize)
+#else
+{-# WARNING fileBlockSize "fileBlockSize: not available on this platform" #-}
+fileBlockSize =


btw, this CPP branch is missing its typesig

Should be fine now.

DanielG · 2018-07-24T11:57:34Z

@hvr what's the status on this? Any reason this never got merged?

llelf · 2018-07-28T15:35:27Z

AFAIU st_blocks is mandatory in POSIX

it's not.

DanielG · 2018-07-29T18:24:11Z

Indeed both st_blocks and st_blksize are XSI options. Thanks @llelf

Bodigrim · 2020-11-14T15:35:22Z

This is blocked until a) either we drop support of older GHCs, b) or agree our stance on conditional APIs, c) fix backward compatibility in other way.

hs-viktor

Looks largely fine, but unless I'm missing something about who sees the #WARNINGs about missing optional structure members, those warnings should go.

System/Posix/Files/Common.hsc

hs-viktor · 2020-11-17T07:06:30Z

System/Posix/Files/Common.hsc

@@ -250,6 +251,10 @@ specialDeviceID  :: FileStatus -> DeviceID
 -- | Size of the file in bytes. If this file is a symbolic link the size is
 -- the length of the pathname it contains.
 fileSize         :: FileStatus -> FileOffset
+-- | Number of block blocks allocated for this file, in units of 512-byte.
+fileBlocks       :: FileStatus -> Maybe CBlkCnt


The types CBlkCnt and CBlkSize below don't appear to be defined with GHC <= 8.0 breaking CI.

Indeed these were introduced conditionally in GHC 8.2 and this PR can probably use the corresponding macros to detect their presence. The new feature can then only be offered if the stat structure has the requisite fields (already done) AND the types are available.

Ah, right. Not sure how we missed that before. I'll add some more GLASGOW_HASKELL version checking then.

This CPP guard is not sufficient. GHC 8.2 conditionally defines these types. So you need to:

#include "HsBaseConfig.h"

and use the same feature macros.

Which means that the types might in fact be unsupported on some platforms even with GHC >= 8.2. However, when that's the case we could define them to be new types for Void in that case, and then the functions can return Maybe X -- X ~ Void, which can only be instantiated as Nothing.

I tried that, but that doesn't work because HsBaseConfig.h defines a bunch of conflicting defines like PACKAGE_NAME etc. which we also have in HsUnix.h.

I'm pretty sure that if the st_ fields exist the corresponding types must also be defined in the system headers so there shouldn't be any need to check for the types and struct fields separately.

then the functions can return Maybe X -- X ~ Void

I thought about that too, but then the API is conditional depending on platform factors again :)

I'm pretty sure that if the st_ fields exist the corresponding types must also be defined

Looks like these didn't always exist, at least according to https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/sys_stat.h.html

Issue 5

[...]
The type of st_blksize is changed from long to blksize_t; the type of st_blocks is changed from long to blkcnt_t.

hs-viktor · 2020-11-17T17:56:53Z

This is blocked until a) either we drop support of older GHCs, b) or agree our stance on conditional APIs, c) fix backward compatibility in other way.

I am not sure why "older GHCs" come into play here. Sure we need at least GHC 8.2 for the types, but since in any case the member fields of "stat" are not universally present, and the PR now returns Nothing on platforms where they're missing, it can (and should) also return Nothing with older GHCs, so the API will be the same API for all systems, but on some systems the return value will always be Nothing. I think that's acceptable.

In other words, given GHC >= 8.2, and a platform that has these fields in struct stat the API will return Just <value> for these fields, and otherwise Nothing. I think this PR can move forward mostly as-is, with appropriate CPP conditionals.

DanielG

it can (and should) also return Nothing with older GHCs

No it can't because we can't mention the new types since they don't exist on old GHCs yet.

I think there is simply no (good) way to avoid having a conditional API depending on the GHC version. We can however still offer the same API regardless of the availability of the two struct fields. I think that's acceptable.

System/Posix/Files/Common.hsc

DanielG · 2020-11-17T18:47:55Z

System/Posix/Files/Common.hsc

@@ -250,6 +251,10 @@ specialDeviceID  :: FileStatus -> DeviceID
 -- | Size of the file in bytes. If this file is a symbolic link the size is
 -- the length of the pathname it contains.
 fileSize         :: FileStatus -> FileOffset
+-- | Number of block blocks allocated for this file, in units of 512-byte.
+fileBlocks       :: FileStatus -> Maybe CBlkCnt


Ah, right. Not sure how we missed that before. I'll add some more GLASGOW_HASKELL version checking then.

DanielG · 2020-11-17T18:58:31Z

Changes:

Decided to use MIN_VERSION_base instead of GLASGOW_HASKELL as a more direct check for the CBlk* types.

DanielG · 2020-11-17T19:00:44Z

Changes:

Fix typo in fileBlocks docs

hs-viktor · 2021-01-19T03:10:12Z

Any interest in rebasing this on master, which now has a more comprehensive CI?
This looks rather close now. I'd like to see whether it is possible to get sensible behaviour with GHC 8.0, with the functions returning Nothing :: Maybe SomeIntegralType for some suitable choice of SomeIntegralType.

hs-viktor · 2021-02-14T00:50:12Z

Please rebase, now that #175 is merged.

DanielG · 2021-02-14T01:02:37Z

Sure, thanks for the ping :)

System/Posix/Files.hsc

DanielG · 2021-02-14T02:08:16Z

Changes:

Remove #ifdefs for MIN_VERSION_base

vdukhovni

Looks ready overall, just curious what others make of my strictness question?

System/Posix/Files/Common.hsc

hs-viktor

With FreeBSD passing CI after initially timing out in some sort of odd virtualbox error, this appears to be ready. I'm satisfied with the Just/unsafePerformIO order rationale by @Bodigrim, so this is ready to go. So I'm approving and merging this PR.

DanielG · 2021-02-15T03:03:38Z

🎉 This has to be my longest running PR by far, hehe, only took a measly 4.25 years to get merged, thanks everyone :D

vdukhovni · 2021-02-15T04:45:56Z

Congratulations! Now that we've sorted out the CI, and dropped support for old GHC versions that can make it difficult to adopt new features, it should be possible to address the backlog without similar delays.

DanielG force-pushed the more-stat-fields branch from 2403c59 to b2a4c24 Compare December 12, 2016 20:04

hvr reviewed Dec 12, 2016

View reviewed changes

hvr added this to the 2.7.3.0 milestone Apr 30, 2017

DanielG mentioned this pull request Jul 24, 2018

Add BlockSize and BlocksCount fields for FileStatus #46

Closed

DanielG force-pushed the more-stat-fields branch 2 times, most recently from f2f5c65 to 8142f27 Compare July 29, 2018 18:34

hvr mentioned this pull request Nov 19, 2018

Add missing blockSize and blockCount accessors to System.Posix.Files. #123

Closed

Bodigrim added the status: blocked label Nov 14, 2020

emilypi mentioned this pull request Nov 16, 2020

Release Planning: unix-2.8.0.0 (GHC 9.6.x) #164

Closed

5 tasks

hs-viktor requested changes Nov 17, 2020

View reviewed changes

System/Posix/Files/Common.hsc Outdated Show resolved Hide resolved

System/Posix/Files/Common.hsc Outdated Show resolved Hide resolved

hs-viktor reviewed Nov 17, 2020

View reviewed changes

DanielG force-pushed the more-stat-fields branch 2 times, most recently from 11b3a8a to ef0cc33 Compare November 17, 2020 18:48

DanielG commented Nov 17, 2020

View reviewed changes

DanielG force-pushed the more-stat-fields branch from ef0cc33 to 2e59464 Compare November 17, 2020 18:57

DanielG force-pushed the more-stat-fields branch from 2e59464 to d7ea7ed Compare November 17, 2020 19:00

DanielG force-pushed the more-stat-fields branch from d7ea7ed to d60d7bc Compare February 14, 2021 01:02

Bodigrim reviewed Feb 14, 2021

View reviewed changes

System/Posix/Files.hsc Outdated Show resolved Hide resolved

Add accessors for st_blocks and st_blksize

edd6b33

DanielG force-pushed the more-stat-fields branch from d60d7bc to edd6b33 Compare February 14, 2021 02:07

Bodigrim approved these changes Feb 14, 2021

View reviewed changes

vdukhovni reviewed Feb 14, 2021

View reviewed changes

System/Posix/Files/Common.hsc Show resolved Hide resolved

System/Posix/Files/Common.hsc Show resolved Hide resolved

hs-viktor approved these changes Feb 15, 2021

View reviewed changes

hs-viktor merged commit 9c1619b into haskell:master Feb 15, 2021

Bodigrim modified the milestones: 2.7.3.0, 2.8.0.0 Apr 20, 2022

Add accessors for st_blocks and st_blksize #78

Add accessors for st_blocks and st_blksize #78

Uh oh!

Conversation

DanielG commented Nov 2, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

hvr commented Nov 2, 2016

Uh oh!

DanielG commented Dec 12, 2016

Uh oh!

hvr commented Dec 12, 2016

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

DanielG commented Jul 24, 2018

Uh oh!

llelf commented Jul 28, 2018

Uh oh!

DanielG commented Jul 29, 2018

Uh oh!

Bodigrim commented Nov 14, 2020

Uh oh!

hs-viktor left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Issue 5

Uh oh!

hs-viktor commented Nov 17, 2020

Uh oh!

DanielG left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

DanielG commented Nov 17, 2020

Uh oh!

DanielG commented Nov 17, 2020

Uh oh!

hs-viktor commented Jan 19, 2021

Uh oh!

hs-viktor commented Feb 14, 2021

Uh oh!

DanielG commented Feb 14, 2021

Uh oh!

Uh oh!

DanielG commented Feb 14, 2021

Uh oh!

vdukhovni left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

hs-viktor left a comment

Choose a reason for hiding this comment

Uh oh!

DanielG commented Feb 15, 2021

Uh oh!

vdukhovni commented Feb 15, 2021

Uh oh!

DanielG commented Nov 2, 2016 •

edited

Loading