Re: [PATCH 4/4] fs/hfsplus: Mark error strings for translation

2023-04-21 Thread Daniel Kiper
On Fri, Apr 21, 2023 at 12:03:52AM +0200, Vladimir 'phcoder' Serbinenko wrote:
> Is there any reason to translate those strings? They refer to deep
> HFS+ structures and problems and to common people they are meaningless
> even in their language. And someone in IT is likely to understand
> those concepts in English. I wouldn't know a word for "extent" in my
> native language. And those errors are very rare as they indicate a

Yeah, I know what you mean... :-)

> corrupted FS in the files GRUB cares about. OTOH translating them
> requires a lot of effort on translator's side for little benefit. They
> were not forgotten. They were not marked for translation on purpose.
> If we mark all errors for translation we will make our already large
> list of strings 10x larger and 90% of it are hyper-obscure error
> conditions. Error conditions are translated only if they are likely to
> occur and indicate a resolution direction like "linux command needs to
> be run before initrd command"

This patch has been suggested by me because I thought the lack of N_()
was a mistake. Your comments shed some light to this. Though I still
think some messages, e.g. "not a HFS+ filesystem", could be translated.

Daniel

> On Thu, Apr 20, 2023 at 8:00 PM Lidong Chen  wrote:
> >
> > Signed-off-by: Lidong Chen 
> > ---
> >  grub-core/fs/hfsplus.c | 30 +++---
> >  1 file changed, 15 insertions(+), 15 deletions(-)
> >
> > diff --git a/grub-core/fs/hfsplus.c b/grub-core/fs/hfsplus.c
> > index cf13e8a63..f91af1477 100644
> > --- a/grub-core/fs/hfsplus.c
> > +++ b/grub-core/fs/hfsplus.c
> > @@ -179,7 +179,7 @@ grub_hfsplus_read_block (grub_fshelp_node_t node, 
> > grub_disk_addr_t fileblock)
> >if (node->fileid == GRUB_HFSPLUS_FILEID_OVERFLOW)
> > {
> >   grub_error (GRUB_ERR_READ_ERROR,
> > - "extra extents found in an extend overflow file");
> > + N_("extra extents found in an extend overflow file"));
> >   break;
> > }
> >
> > @@ -190,7 +190,7 @@ grub_hfsplus_read_block (grub_fshelp_node_t node, 
> > grub_disk_addr_t fileblock)
> >if (!node->data->extoverflow_tree_ready)
> > {
> >   grub_error (GRUB_ERR_BAD_FS,
> > - "attempted to read extent overflow tree before 
> > loading");
> > + N_("attempted to read extent overflow tree before 
> > loading"));
> >   break;
> > }
> >
> > @@ -205,8 +205,8 @@ grub_hfsplus_read_block (grub_fshelp_node_t node, 
> > grub_disk_addr_t fileblock)
> >   || !nnode)
> > {
> >   grub_error (GRUB_ERR_READ_ERROR,
> > - "no block found for the file id 0x%x and the block"
> > - " offset 0x%" PRIuGRUB_UINT64_T,
> > + N_("no block found for the file id 0x%x and the block"
> > + " offset 0x%" PRIuGRUB_UINT64_T),
> >   node->fileid, fileblock);
> >   break;
> > }
> > @@ -277,7 +277,7 @@ grub_hfsplus_mount (grub_disk_t disk)
> >/* See if there's an embedded HFS+ filesystem.  */
> >if (grub_be_to_cpu16 (volheader.hfs.embed_sig) != GRUB_HFSPLUS_MAGIC)
> > {
> > - grub_error (GRUB_ERR_BAD_FS, "not a HFS+ filesystem");
> > + grub_error (GRUB_ERR_BAD_FS, N_("not a HFS+ filesystem"));
> >   goto fail;
> > }
> >
> > @@ -303,7 +303,7 @@ grub_hfsplus_mount (grub_disk_t disk)
> >|| ((volheader.hfsplus.blksize & (volheader.hfsplus.blksize - 1)) != 
> > 0)
> >|| grub_be_to_cpu32 (volheader.hfsplus.blksize) < 
> > GRUB_DISK_SECTOR_SIZE)
> >  {
> > -  grub_error (GRUB_ERR_BAD_FS, "not a HFS+ filesystem");
> > +  grub_error (GRUB_ERR_BAD_FS, N_("not a HFS+ filesystem"));
> >goto fail;
> >  }
> >
> > @@ -358,7 +358,7 @@ grub_hfsplus_mount (grub_disk_t disk)
> >
> >if (data->catalog_tree.nodesize < 2)
> >  {
> > -  grub_error (GRUB_ERR_BAD_FS, "invalid catalog node size");
> > +  grub_error (GRUB_ERR_BAD_FS, N_("invalid catalog node size"));
> >goto fail;
> >  }
> >
> > @@ -378,7 +378,7 @@ grub_hfsplus_mount (grub_disk_t disk)
> >
> >if (data->extoverflow_tree.nodesize < 2)
> >  {
> > -  grub_error (GRUB_ERR_BAD_FS, "invalid extents overflow node size");
> > +  grub_error (GRUB_ERR_BAD_FS, N_("invalid extents overflow node 
> > size"));
> >goto fail;
> >  }
> >
> > @@ -406,7 +406,7 @@ grub_hfsplus_mount (grub_disk_t disk)
> >   fail:
> >
> >if (grub_errno == GRUB_ERR_OUT_OF_RANGE)
> > -grub_error (GRUB_ERR_BAD_FS, "not a HFS+ filesystem");
> > +grub_error (GRUB_ERR_BAD_FS, N_("not a HFS+ filesystem"));
> >
> >grub_free (data);
> >return 0;
> > @@ -550,7 +550,7 @@ grub_hfsplus_btree_iterate_node (struct 
> > grub_hfsplus_btree *btree,
> >
> >if (node_count && first_node->next == saved_node)
> > {
> > - grub_error (GRUB_ERR_B

Re: [PATCH 4/4] fs/hfsplus: Mark error strings for translation

2023-04-21 Thread Lidong Chen


> On Apr 21, 2023, at 6:10 AM, Daniel Kiper  wrote:
> 
> On Fri, Apr 21, 2023 at 12:03:52AM +0200, Vladimir 'phcoder' Serbinenko wrote:
>> Is there any reason to translate those strings? They refer to deep
>> HFS+ structures and problems and to common people they are meaningless
>> even in their language. And someone in IT is likely to understand
>> those concepts in English. I wouldn't know a word for "extent" in my
>> native language. And those errors are very rare as they indicate a
> 
> Yeah, I know what you mean... :-)
> 
>> corrupted FS in the files GRUB cares about. OTOH translating them
>> requires a lot of effort on translator's side for little benefit. They
>> were not forgotten. They were not marked for translation on purpose.
>> If we mark all errors for translation we will make our already large
>> list of strings 10x larger and 90% of it are hyper-obscure error
>> conditions. Error conditions are translated only if they are likely to
>> occur and indicate a resolution direction like "linux command needs to
>> be run before initrd command"
> 
> This patch has been suggested by me because I thought the lack of N_()
> was a mistake. Your comments shed some light to this. Though I still
> think some messages, e.g. "not a HFS+ filesystem", could be translated.

 I can revert the changes, except the general messages as mentioned by Daniel.

Thanks,
Lidong

> 
> Daniel
> 
>> On Thu, Apr 20, 2023 at 8:00 PM Lidong Chen  wrote:
>>> 
>>> Signed-off-by: Lidong Chen 
>>> ---
>>> grub-core/fs/hfsplus.c | 30 +++---
>>> 1 file changed, 15 insertions(+), 15 deletions(-)
>>> 
>>> diff --git a/grub-core/fs/hfsplus.c b/grub-core/fs/hfsplus.c
>>> index cf13e8a63..f91af1477 100644
>>> --- a/grub-core/fs/hfsplus.c
>>> +++ b/grub-core/fs/hfsplus.c
>>> @@ -179,7 +179,7 @@ grub_hfsplus_read_block (grub_fshelp_node_t node, 
>>> grub_disk_addr_t fileblock)
>>>   if (node->fileid == GRUB_HFSPLUS_FILEID_OVERFLOW)
>>>{
>>>  grub_error (GRUB_ERR_READ_ERROR,
>>> - "extra extents found in an extend overflow file");
>>> + N_("extra extents found in an extend overflow file"));
>>>  break;
>>>}
>>> 
>>> @@ -190,7 +190,7 @@ grub_hfsplus_read_block (grub_fshelp_node_t node, 
>>> grub_disk_addr_t fileblock)
>>>   if (!node->data->extoverflow_tree_ready)
>>>{
>>>  grub_error (GRUB_ERR_BAD_FS,
>>> - "attempted to read extent overflow tree before 
>>> loading");
>>> + N_("attempted to read extent overflow tree before 
>>> loading"));
>>>  break;
>>>}
>>> 
>>> @@ -205,8 +205,8 @@ grub_hfsplus_read_block (grub_fshelp_node_t node, 
>>> grub_disk_addr_t fileblock)
>>>  || !nnode)
>>>{
>>>  grub_error (GRUB_ERR_READ_ERROR,
>>> - "no block found for the file id 0x%x and the block"
>>> - " offset 0x%" PRIuGRUB_UINT64_T,
>>> + N_("no block found for the file id 0x%x and the block"
>>> + " offset 0x%" PRIuGRUB_UINT64_T),
>>>  node->fileid, fileblock);
>>>  break;
>>>}
>>> @@ -277,7 +277,7 @@ grub_hfsplus_mount (grub_disk_t disk)
>>>   /* See if there's an embedded HFS+ filesystem.  */
>>>   if (grub_be_to_cpu16 (volheader.hfs.embed_sig) != GRUB_HFSPLUS_MAGIC)
>>>{
>>> - grub_error (GRUB_ERR_BAD_FS, "not a HFS+ filesystem");
>>> + grub_error (GRUB_ERR_BAD_FS, N_("not a HFS+ filesystem"));
>>>  goto fail;
>>>}
>>> 
>>> @@ -303,7 +303,7 @@ grub_hfsplus_mount (grub_disk_t disk)
>>>   || ((volheader.hfsplus.blksize & (volheader.hfsplus.blksize - 1)) != 
>>> 0)
>>>   || grub_be_to_cpu32 (volheader.hfsplus.blksize) < 
>>> GRUB_DISK_SECTOR_SIZE)
>>> {
>>> -  grub_error (GRUB_ERR_BAD_FS, "not a HFS+ filesystem");
>>> +  grub_error (GRUB_ERR_BAD_FS, N_("not a HFS+ filesystem"));
>>>   goto fail;
>>> }
>>> 
>>> @@ -358,7 +358,7 @@ grub_hfsplus_mount (grub_disk_t disk)
>>> 
>>>   if (data->catalog_tree.nodesize < 2)
>>> {
>>> -  grub_error (GRUB_ERR_BAD_FS, "invalid catalog node size");
>>> +  grub_error (GRUB_ERR_BAD_FS, N_("invalid catalog node size"));
>>>   goto fail;
>>> }
>>> 
>>> @@ -378,7 +378,7 @@ grub_hfsplus_mount (grub_disk_t disk)
>>> 
>>>   if (data->extoverflow_tree.nodesize < 2)
>>> {
>>> -  grub_error (GRUB_ERR_BAD_FS, "invalid extents overflow node size");
>>> +  grub_error (GRUB_ERR_BAD_FS, N_("invalid extents overflow node 
>>> size"));
>>>   goto fail;
>>> }
>>> 
>>> @@ -406,7 +406,7 @@ grub_hfsplus_mount (grub_disk_t disk)
>>>  fail:
>>> 
>>>   if (grub_errno == GRUB_ERR_OUT_OF_RANGE)
>>> -grub_error (GRUB_ERR_BAD_FS, "not a HFS+ filesystem");
>>> +grub_error (GRUB_ERR_BAD_FS, N_("not a HFS+ filesystem"));
>>> 
>>>   grub_free (data);
>>>   return 0;
>>> @@ -550,7 +550,7 @@ grub_hfsplus_btree_iterat

Installed Fedora 38 (sway) / grub2 on a new HD - boots OK but does not have a menuentry item in grub.cfg ?

2023-04-21 Thread Philip Rhoades via Grub-devel

People,

My Fedora WS has 4 drive bays and I access old boot and other HDs in 
bays 2-4.  I just did a clean install of Fedora Sway to /dev/sda - 
everything is fine and the system boots OK but I am confused:


- cat /etc/os-release

NAME="Fedora Linux"
VERSION="38 (Sway)"
.
PRETTY_NAME="Fedora Linux 38 (Sway)"
ANSI_COLOR="0;38;2;60;110;180"
.
VARIANT="Sway"
VARIANT_ID=sway

- The install picked up old Fedora boot setups which show up in 
/boot/grub2/grub.conf like this:


menuentry 'Fedora Linux 37 (KDE Plasma) (on /dev/sde3)' --class 
gnu-linux --class gnu --class os $menuentry_id_option 
'osprober-gnulinux-/boot/vmlinuz-6.0.7-301.fc37.x86_64--f6f43412-e6ed-4dff-808b-fe013691c3b6' 
{


- but why is there is not a line like this in /boot/grub2/grub.conf for 
Fedora 38 Sway?


I can see that the appropriate other Sway files are in the /boot tree so 
I am guessing that grub2 somehow knows the proper, default boot image - 
even though the default does not have a menuentry in grub.cfg?


A pointer to a link for info would be great . .

Thanks!

Phil.
--
Philip Rhoades

PO Box 896
Cowra  NSW  2794
Australia
E-mail:  p...@pricom.com.au

___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel