Long Filenames conversion

John Tr.
11/21/2006 03:20 pm
I download and tried Offline Explorer Pro to try rip something from ovid.
Before doing so I read in this forum that long filenames are cut and converted to short ones using an algorithm that generated a unique filename.
However I noticed that the filename generated s not always unique
For example this 2 long filenames generated the same short filename resulting in downloading the same files instead of downloading the correct ones.

...../OVIDBOOK[1]/TXTBKBD[1]/DIVISIONA[1]/DIVISIONB[1]/DIVISIONC[1]/CHAPTER[1]
....//OVIDBOOK[1]/TXTBKBD[1]/DIVISIONA[2]/DIVISIONB[1]/DIVISIONC[1]/CHAPTER[1]

The only difference in the filename is A[1] with A[2].
However this seemed not enough to generate a unique short filename.
Is there any solution for this?
Thank you in advance
Best Regards
Oleg Chernavin
11/22/2006 03:22 am
Offline Explorer uses long filenames, but if they exceed 100 symbols, they are cut and an unique number is being added to the end of filename to make sure it will differ from simular URLs.

Best regards,
Oleg Chernavin
MP Staff
John Tr.
11/22/2006 04:02 am
Hi Oleg, thank you for your reply.
What I noticed is that at least in one case the unique number was not unique.
In the example given the 2 links generated the same unique number.
This resulted in downloading again the same files.

This problem appeared only in one book from ovid that had this configuration
...../OVIDBOOK[1]/TXTBKBD[1]/DIVISIONA[1]/DIVISIONB[1]/DIVISIONC[1]/CHAPTER[1]
....//OVIDBOOK[1]/TXTBKBD[1]/DIVISIONA[2]/DIVISIONB[1]/DIVISIONC[1]/CHAPTER[1]

A[1] and A[2] generated the same unique number.Is this normal?
Thank you again!

Oleg Chernavin
11/22/2006 04:06 am
Can you please post full URLs and filenames of them here?

Oleg.
John Tr.
11/22/2006 04:34 am
An example is following:

Both these links
hxxp://gateway.ut.ovid.com/gw1/ovidweb.cgi?targetFrame=1&S=IDNJHKKKCBEMHO00D&Book+Content=S.sh.2.14.15%7c17%7c00000000%7c%7c/CT{06b9ee1beed59419e3e863e77329428570869c27e1523f4b1163e28761ebf259552f94da368611b46686bb0e4bb90578}/OVIDBOOK[1]/TXTBKBD[1]/DIVISIONA[1]/DIVISIONB[1]/DIVISIONC[2]/CHAPTER[1]
hxxp://gateway.ut.ovid.com/gw1/ovidweb.cgi?targetFrame=1&S=IDNJHKKKCBEMHO00D&Book+Content=S.sh.2.14.15%7c17%7c00000000%7c%7c/CT{06b9ee1beed59419e3e863e77329428570869c27e1523f4b1163e28761ebf259552f94da368611b46686bb0e4bb90578}/OVIDBOOK[1]/TXTBKBD[1]/DIVISIONA[2]/DIVISIONB[1]/DIVISIONC[1]/CHAPTER[1]

generated the same filename
ovidweb.cgi@targetFrame=1&S=IDNJHKKKCBEMHO00D&Book+Content=S.sh.2.14.15_257c17_257c00000000_257c_257c_2FCT{06b9ee1bee0A1BE0AD02
John Tr.
11/22/2006 04:36 am
I am sorry for double posting. I would like only to add that this is not the only example.
This appeared many times in the particular books, after the links changed from A[1] to A[2]
The only solution I found for this was to create 2 folders and cutting the download process into to parts , each of this on a different folder.
1 part for A[1] links
and 1 part for A[2] links

Thank you!
Oleg Chernavin
11/22/2006 06:28 am
Yes, this is a very rare case when such calculated ending is the same for both URLs. I am sorry, but I cannot do anything right now.

Oleg.
John Tr.
11/22/2006 07:05 am
Anyway,
Thank you for your answers.
Oleg Chernavin
11/22/2006 07:10 am
You are welcome!

Oleg.