Detect file type by content?

[Little John]

Hi,

I have “inherited” an external hard disk with several files that have no extension in their name. I assume that these are mainly files which were created with Word, Excel or PowerPoint, as well as raster graphics. Do these files have a header or something that allows them to be clearly identified programmatically?

[StarBootics]

Maybe they have a magic number at the beginning you can try to read.

Beside that I don't know what else can be used to identify the files.

Best regards
StarBootics

[jacdelad]

TrID: https://mark0.net/soft-trid-e.html
...if you don't necessarily want to program something yourself.

[ChrisR]

TrID - File Identifier looks like a very interesting tool

[jacdelad]

I've used it several times. It's a bit fiddly, but since we're programmers and its output should be automatically readable it should be possible to create some kind of gui easily.

[AZJIO]

Word, Excel or PowerPoint

This is a zip archive. I previously wrote a program to replace text in Word files, for this I unpacked the file using the zip module, and after the replacement I packed it back and it worked, the files opened.

[Little John]

TrID: https://mark0.net/soft-trid-e.html
...if you don't necessarily want to program something yourself.

That looks very promising. I'll see how far I get with it next week in the office. Many thanks for the tip!
Many thanks also to StarBootics for the link to the magic numbers!

[NicTheQuick]

Under Linux you can simply use the "file" tool to recognize any file. Linux does it that way anyway.
Or you also can try "binwalk" which can also find other files within a binar data blob. This is especially interesting when you want to reverse engineer file formats or firmware files and stuff like that.

[Little John]

Under Linux you can simply use the "file" tool to recognize any file. Linux does it that way anyway.
Or you also can try "binwalk" which can also find other files within a binar data blob. This is especially interesting when you want to reverse engineer file formats or firmware files and stuff like that.

This is valuable information, thank you. It seems that I am too focused on Windows.

[DarkDragon]

Under Linux you can simply use the "file" tool to recognize any file. Linux does it that way anyway.
Or you also can try "binwalk" which can also find other files within a binar data blob. This is especially interesting when you want to reverse engineer file formats or firmware files and stuff like that.

This is valuable information, thank you. It seems that I am too focused on Windows.

I use file on windows, too, if I have to. Usually within a cygwin, but nowadays you could try WSL/WSL2. There's also MinGW: https://sourceforge.net/projects/mingw/ ... le-5.04-1/

[Little John]

I use file on windows, too, if I have to. Usually within a cygwin, but nowadays you could try WSL/WSL2. There's also MinGW: https://sourceforge.net/projects/mingw/ ... le-5.04-1/

I wanted to try WSL anyway, but had almost forgotten about it. Thanks for reminding me!

[Quin]

I use file on windows, too, if I have to. Usually within a cygwin, but nowadays you could try WSL/WSL2. There's also MinGW: https://sourceforge.net/projects/mingw/ ... le-5.04-1/

I wanted to try WSL anyway, but had almost forgotten about it. Thanks for reminding me!

Believe it or not, all my browsing of the PB forums is done within WSL, using edbrowse.
I have to disable Javascript or the ads make it hang for like 30 seconds on every page load, but that's expected for edbrowse.