udev patch for attributes with trailing spaces in the values

udev patch for attributes with trailing spaces in the values

[Mark Eaton]

[-- Attachment #1: Type: text/plain, Size: 511 bytes --]

The following patch strips off spaces that are often contained at the end of 
SYSFS_ attributes. (ie lots of printers etc do this).
I'm not sure if this should actually be added to udev but I would be curious 
to hear peoples thoughts as I find it extremly annoying and pointless to have 
to hexdump the sysfs values to check if there are trailing spaces to write 
the various rules.


--
Mark Eaton
Senior Software Engineer
VASCO The Authentication Company
+67 3511 5404
+67 (0) 403 384 886
meaton@vasco.com



[-- Attachment #2: spaces.patch --]
[-- Type: text/x-diff, Size: 503 bytes --]

diff -Nru udev-015-orig/namedev.c udev-015/namedev.c
--- udev-015-orig/namedev.c	2004-01-23 09:28:36.000000000 +1000
+++ udev-015/namedev.c	2004-01-30 08:59:10.000000000 +1000
@@ -427,7 +427,10 @@
 label_found:
 	c = tmpattr->value + strlen(tmpattr->value)-1;
 	if (*c == '\n')
-		*c = 0x00;
+		*c-- = 0x00;
+	while (*c == ' ')
+		*c-- = 0x00;
+	
 	dbg("compare attribute '%s' value '%s' with '%s'",
 		  pair->file, tmpattr->value, pair->value);
 	if (strcmp_pattern(pair->value, tmpattr->value) != 0)

Re: udev patch for attributes with trailing spaces in the values

[Kay Sievers]

On Fri, Jan 30, 2004 at 10:14:51AM +1000, Mark Eaton wrote:
> The following patch strips off spaces that are often contained at the end of 
> SYSFS_ attributes. (ie lots of printers etc do this).
> I'm not sure if this should actually be added to udev but I would be curious 
> to hear peoples thoughts as I find it extremly annoying and pointless to have 
> to hexdump the sysfs values to check if there are trailing spaces to write 
> the various rules.

You may use udevinfo that prints the ready formated attribute for you
or use the pattern matching if you don't want to count the number of
spaces. :)



> diff -Nru udev-015-orig/namedev.c udev-015/namedev.c
> --- udev-015-orig/namedev.c	2004-01-23 09:28:36.000000000 +1000
> +++ udev-015/namedev.c	2004-01-30 08:59:10.000000000 +1000
> @@ -427,7 +427,10 @@
>  label_found:
>  	c = tmpattr->value + strlen(tmpattr->value)-1;
>  	if (*c = '\n')
> -		*c = 0x00;
> +		*c-- = 0x00;

Huh, what if the value is only the '\n'?

> +	while (*c = ' ')
> +		*c-- = 0x00;


thanks,
Kay


-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev patch for attributes with trailing spaces in the values

[Darren Salt]

[-- Attachment #1: Type: text/plain, Size: 1665 bytes --]

I demand that Kay Sievers may or may not have written...

> On Fri, Jan 30, 2004 at 10:14:51AM +1000, Mark Eaton wrote:
>> The following patch strips off spaces that are often contained at the end
>> of SYSFS_ attributes. (ie lots of printers etc do this). I'm not sure if
>> this should actually be added to udev but I would be curious to hear
>> peoples thoughts as I find it extremly annoying and pointless to have to
>> hexdump the sysfs values to check if there are trailing spaces to write
>> the various rules.

> You may use udevinfo that prints the ready formated attribute for you or
> use the pattern matching if you don't want to count the number of spaces.
> :)

I recently hit this problem when adding a device name (mainly "because I
can") for a CF card reader. Initially, I wasn't aware that there /were/ any
trailing spaces (and I hadn't noticed udevinfo), then I looked again and went
hexdumping...

>> diff -Nru udev-015-orig/namedev.c udev-015/namedev.c
>> --- udev-015-orig/namedev.c 2004-01-23 09:28:36.000000000 +1000
>> +++ udev-015/namedev.c 2004-01-30 08:59:10.000000000 +1000
>> @@ -427,7 +427,10 @@
>>  label_found:
>>   c = tmpattr->value + strlen(tmpattr->value)-1;
>>   if (*c == '\n')
>> -  *c = 0x00;
>> +  *c-- = 0x00;

> Huh, what if the value is only the '\n'?

Fun and games, of course ;-)

My independently-written version is attached.

-- 
| Darren Salt | nr. Ashington, | d youmustbejoking,demon,co,uk
| Debian,     | Northumberland | s zap,tartarus,org
| RISC OS     | Toon Army      | @
|   When replying, "news"->"ds", "cu"->"co.uk", and validate

By protracting life, we do not deduct one jot from the duration of death.

[-- Attachment #2: trailing-space.patch --]
[-- Type: text/plain, Size: 531 bytes --]

--- udev-016/namedev.c~	2004-01-31 15:36:18.000000000 +0000
+++ udev-016/namedev.c	2004-02-11 15:09:06.000000000 +0000
@@ -422,9 +422,9 @@
 	return -ENODEV;
 
 label_found:
-	c = tmpattr->value + strlen(tmpattr->value)-1;
-	if (*c == '\n')
-		*c = 0x00;
+	c = tmpattr->value + strlen(tmpattr->value);
+	while (--c >= tmpattr->value && isspace (*c))
+		*c = 0;
 	dbg("compare attribute '%s' value '%s' with '%s'",
 		  pair->file, tmpattr->value, pair->value);
 	if (strcmp_pattern(pair->value, tmpattr->value) != 0)