- From: Juan Carlos Cruellas <cruellas@ac.upc.edu>
- Date: Tue, 12 Jun 2007 14:14:28 +0200
- To: XMLSec <public-xmlsec-maintwg@w3.org>
Dear all,
I have started reviewing the proposed annex text and I have found
something in the text and also in some of the examples (which I have not
been able to completely process).
1. On the text:
in 2.C the text reads:
if the input buffer begins with a prefix of "/../" or "/..", where ".."
is a complete path segment, then replace that prefix with "/" in the
input buffer and if also the output buffer is empty, last segment in the
output buffer equals "../" or "..", where ".." is a complete path
segment, then append ".." or "/.." for the latter case respectively to
the output buffer else remove the last segment and its preceding "/" (if
any) from the output buffer; otherwise,
I think that the behaviour for the case "output buffer is empty" is left
undefined. There are three situations:
1. output buffer is empty
2. out buffer equals "../"
3. out buffer equals ".."
But there are only two ways of obtaining the output:
"then append ".." or "/.." for the latter case respectively to the
output buffer".
By reading this I do not know what I have to add to the output buffer in
case 1. (out buffer is empty).
2. As I said, I have gone through some of the examples (I did not have
time to go through all of them, appologies). And in some of them I have
got different results from the ones that appear in the annex...Of
course, I am not completely sure whether I have missread something...but
most of the discrepancies seem to appear because in steps B and C there
is a "replace that prefix with "/" in the input buffer that I always
include but in the current text in the annex seems to dissapear.
Below follows the details of the results that I have obtained following
my understanding of the annex step by step.
Step 0 is the substitution of "//" by "/".
Step 1 onwards are the steps for processing dots. The letter after the
step number reflects the case that according to my understanding applies
according to the start of the input buffer.
STEP OUTPUT INPUT
no/.././/pseudo-netpath/seg/file.ext
0.
no/.././pseudo-netpath/seg/file.ext
1.E no /.././pseudo-netpath/seg/file.ext
1.C "" /./pseudo-netpath/seg/file.ext
1.B "" /pseudo-netpath/seg/file.ext
1.E /pseudo-netpath /seg/file.ext
1.E /pseudo-netpath/seg /file.ext
1.E /pseudo-netpath/seg/file.ext
**DIFFERENT RESULT IN THE ANNEX**
STEP OUTPUT INPUT
no/..//.///pseudo-netpath/seg/file.ext
0.
no/.././pseudo-netpath/seg/file.ext
1.E no /.././pseudo-netpath/seg/file.ext
1.C "" /./pseudo-netpath/seg/file.ext
1.B "" /pseudo-netpath/seg/file.ext
1.E /pseudo-netpath /seg/file.ext
1.E /pseudo-netpath/seg /file.ext
1.E /pseudo-netpath/seg/file.ext
**DIFFERENT RESULT IN THE ANNEX**
STEP OUTPUT INPUT
yes/no//..//.///pseudo-netpath/seg/file.ext
0.
yes/no/.././pseudo-netpath/seg/file.ext
1.E yes
/no/.././pseudo-netpath/seg/file.ext
2.E yes/no /.././pseudo-netpath/seg/file.ext
3.C yes /./pseudo-netpath/seg/file.ext
4.B yes /pseudo-netpath/seg/file.ext
5.E yes/pseudo-netpath /seg/file.ext
6.E yes/pseudo-netpath/seg /file.ext
7.E yes/pseudo-netpath/seg/file.ext
*** RESULT AS IN THE ANNEX ***
STEP OUTPUT INPUT
no/../yes
1.E no /../yes
2.C "" /yes
3.E /yes
**DIFFERENT RESULT IN THE ANNEX**
Please note that, if I have not missunderstood the text, in step 2, case
C of the annex applies and that it reads "replace that prefix with "/"
in the input buffer", so in the input buffer we have "/yes", not "yes",
and in the last step case E applies that instructs to keep the initial "/".
STEP OUTPUT INPUT
no/../yes/
1.E no /../yes/
2.C "" /yes/
3.E /yes /
4.E /yes/
**DIFFERENT RESULT IN THE ANNEX**
As before
STEP OUTPUT INPUT
no/../yes/no/..
1.E no /../yes/no/..
2.C "" /yes/no/..
3.E /yes /no/..
4.E /yes/no /..
5.c /yes /
6.E /yes/
**DIFFERENT RESULT IN THE ANNEX**
As before
STEP OUTPUT INPUT
../../no/../..
1.A ../ ../no/../..
2.A ../../ no/../..
3.E ../../no /../..
4.C ../.. /..
5.C ../../.. /
6.E ../../../
STEP OUTPUT INPUT
no/../..
1.E no /../..
2.C "" /..
3.C .. OR /.. ? /
4.E ../ OR /../ ?
I couldn't go any further... I will keep testing after the meeting.
Regards
Juan Carlos.
Received on Tuesday, 12 June 2007 12:14:46 UTC