For files, the extension is just a sign to help identify the nature of the file.
Like the file name, the extension can use any name and can be modified at will. In addition to the early file extensions (EXE, COM, BAT, SYS) that are formulated by the operating system and established by convention (TXT), the generation of extensions can be roughly divided into the following situations: In the application process of computers, many things are produced Different types of files. For a programmer, if the software he writes needs to generate a file in a new format, he can arbitrarily define an extension for the file. Maybe this software is very successful and has spawned many related programs. For compatibility, this extension has become a patent for this type of file. Such as RAR, CCD (Clone CD) files. Another case is the extension specified according to a certain standard, such as: JPG, MPG. With the extension, you can specify a certain file to process (called an association in Windows) when targeting a certain type of file, which is indeed much more convenient. So is the extension the only way to determine the nature of a file? The answer is no. For example, DAT file is an extension that is widely used, from VCD discs to many data files. Cuteftp’s site management information also uses it as an extension (Site.DAT, only keep this when backing up Cuteftp file will be fine). Obviously, no software can open so many DAT files in one place. In the past during the DOS era, in order to prevent file viruses, someone modified the IO.SYS MSDOS.SYS file to change the extension of the system's default execution file. For example, change EXE to 1XE, COM to C0M, etc. In this way, normal files with EXE extension cannot be executed on the modified system, and the only way is to change the extension to 1XE. In this way, all executable files in the system must change their identities and become 1XE, C0M and the like. The starting point is that when the virus cannot find the EXE or COM file during the infection process and cannot complete the infection, the system will be safe. But in fact, his system may still be infected by viruses, because some viruses infect files based on file header marks. Moreover, this method is cumbersome and very unrealistic to apply. For example, when a program needs to call another program, the file cannot be found because the extension has changed. The only way to make this program work properly is Treat this program like the MSDOS.SYS file and change the name of the called file (somewhat similar to the current Chinese version). If there are dozens or hundreds of such files in the system... As mentioned above, the file header is mentioned. When MS defines an executable file, it allows two types of files to be executed by the system, that is, COM and EXE. The difference between these two types of files is that COM files are completely composed of compiled code. Execution instructions and data are mixed and stored together, and they are separated by jump instructions (in fact, COM can be regarded as an EXE file with only one segment. Due to the search Because of the address, the size of each segment is at most 64K, so the vast majority of COM files are smaller than 64K. Why do we say most, because some techniques can make COM break through the 64K limit, such as some of Norton's tool programs. What skills, don’t ask me, I don’t know either). The EXE file consists of data segments and code segments. In order to distinguish them, a file header is generated by the compiler at the front of the file. This file header contains the information required by the EXE itself when executing, such as the addresses of the code and data segments, the minimum memory required, etc. More importantly, the file header contains a flag that distinguishes EXE and COM files! When we enter a command on the command line, the system will first search for a matching command in the internal command. If not, it will search for BAT COM and EXE files with the same name in the current path and system path. Once found, it will be used by COMMAND.COM Load it and execute it.
As mentioned above, EXE and COM have different formats. So if you change an EXE file to a COM extension or vice versa, if you still process it according to the original rules, you will definitely make an error, but in fact it will not. The reason is that the system does not completely define files based on extensions. When a file is loaded by COMMAND.COM, it first looks for the EXE file mark. If it exists, it will be executed according to EXE rules. If it does not exist, it will be regarded as a COM file. If you are interested, you can do an experiment, including changing a text document to COM or EXE, and see how the system reacts (please test under DOS, especially not under the command prompt of the Win NT/2K/XP series) , since it is a DOS virtual machine). In the EXE file, this mark is the first two bytes of the file. The content is 4D 5A in hexadecimal. The converted ASCII code is MZ. Similarly, many types of files now have file headers, such as all audio and video files, formatted document files, etc. When processing them, the application also uses the flags here to determine the file type and the information of the file itself, such as the compression format, size, index, etc. of audio and video files. Below are the symbols of some common files. I only found a few. You can collect more information yourself. The numbers are in hexadecimal, and the corresponding ASCII characters are in brackets. Executable file EXE: 4D 5A (MZ) Zip file: 50 4B (PK) RAR file: 52 61 72 21 (Rar!) BMP file: 42 4D (BM) When encountering an unknown file, or suspecting the file's When it comes to properties, it's definitely a good idea to look at its file header. Likewise, changing a file's extension does not change its nature. Even if you change a JPG file to AAA, ACDSEE can still view it. So my conclusion is: the extension is defined for the convenience of practical files and has nothing to do with the nature of the file