Detect charset of file dynamically in c++ - c++

I am trying to read a file which may have any charset/codePage, but I don't which locale to set in order to read the file correctly.
Below is my code snippet in which I am trying to read a file having charset as windows-1256, but I want to get the charset dynamically from the file being read so that I can set the locale accordingly.
std::wifstream input{ filename.c_str() };
std::wstring content{ std::istreambuf_iterator<wchar_t>(input1), std::istreambuf_iterator<wchar_t>() };
input.imbue(std::locale(".1256"));
contents = ws2s(content); // Convert wstring to CString

In general, this is impossible to do accurately using the content of a plain text file alone. Usually you should rely on some external information. For example, if the file was downloaded with HTTP, the encoding should be received within a response header.
Some files may contain information about the encoding as specified by the file format. XML for example: <?xml version="1.0" encoding="XXX"?>.
Unicode encodings can be detected if the file starts with a Byte Order Mark - which is optional.
You can usually assume that the encoding uses a wide character if the file contains a zero byte - which would represent the string terminator as a narrow character - before the end of the file. Likewise if you find two consecutive zeroes aligned to a 2 byte boundary (before the end), then the encoding is probably 4 bytes wide.
Other than that, you could try to guess the encoding based on the frequency of certain characters. This can have some unintended consequences.

Let me be blunt and say: you can't
Let me qualify that: a file is simply tons of 0's and 1's stuck on your disk. A charset is a way to interpret these 0's and 1's. You have to provide the information on how to interpret them, namely, by specifying a charset.
A typical way of doing that is by writing a header to specify the charset.
This is a html header
<head>
<title>Page Title</title>
<meta charset="UTF-8">
</head>
As you can see, the charset must be specified one way or another.
Once in a while, you do see some rogue application guessing a charset, they often do so with some heuristics on the distribution of bytes, but that is not reliable and often results in gibberish.
As a side note, try use UTF-8 everywhere, the others are, to put it lightly, messy.

Related

C++: Problem of Korean alphabet encoding in text file write process with std::ofstream

I have a code for save the log as a text file.
It usually works well, but I found a case where doesn't work:
{Id": "testman", "ip": "192.168.1.1", "target": "?뚯뒪??exe", "desc": "?덈뀞諛⑷??뚯슂"}
My code is a simple logic that saves the log string as a text file.
My code was works well when log is English, but there is a problem when log is Korean language.
After checking through various experiments, it was confirmed that Korean language would not problem if the file could be saved as utf-8 format.
I think, if Korean language is included in log string, c++ is basically saved as ANSI format.
This is my c++ code:
string logfilePath = {path};
log = "{\Id\": \"testman\", \"ip\": \"192.168.1.1\", \"target\": \"테스트.exe\", \"desc\": \"안녕방가워요\"}";
ofstream output(logFilePath, ios::app);
output << log << endl;
output.close();
Is there a way to save log files as uft-8 or any other good way?
Please give me some advice.
You could set UTF-8 in File->Advanced Save Options.
If you do not find it, you could add Advanced Save Options in Tools->Customize->Commands->Add Command..->File.
TDLR: write 0xefbbbf (3-bytes UTF-8 BOM) in the beginning of the file before writing out your string.
One of the hints that text viewer software use to determine if the file should be shown in the Unicode format is something called the Byte Order Marker (or BOM for short). It is basically a series of bytes in the beginning of a stream of text that specifies the encoding and endianness of the text string. For UTF-8 it is these three bytes 0xEF 0xBB 0xBF.
You can experiment with this by opening notepad, writing a single character and saving file in the ANSI format. Then look at the size of file in bytes. It will be 1 byte. Now open the file and save it in UTF-8 and look at the size of file again. It will 4 bytes that is three bytes for the BOM and one byte for the single character you put in there. You can confirm this by viewing both files in some hex editor.
That being said, you may need to insert these bytes to your files before writing your string to them. So why UTF-8? you may ask, well, it depends on the encoding the original string is encoded in (your std::string log) which in this case it is an string literal written in a source file whose encoding is (most likely) UTF-8. Therefor the bytes that build up the string are made according to this encoding and are put into your executable.
note that std::string can contain Unicode string, it just can't make sense of it. For example it reports its length wrong. But it can be used to carry Unicode string around fine.

read file of unknown encoding c++

I want to read the contents of log file which is generated by msiexec. The encoding type of the log file is UCS-2 LE BOM(not sure how this encoding type used while generating log file).
When I read the content of this file using below code, I am getting non ascii characters in the string.
std::string errMsg;
std::ifstream ifs("install.log");
for (std::string line; std::getline(ifs, line); /**/)
{
errMsg.append(line);
}
Is there any way to read a file of any encoding and convert to ANSI using C++17?
read file of unknown encoding c++
You can read any data if you use binary mode. Binary mode will not treat null character as end of string and doesn't transform end of line sequences from system specific formats.
and convert to ANSI
This is like asking how to translate a language that you don't understand. It's not possible in general. There is no way to convert unknown encoding to another encoding. You must know the source encoding to achieve that. You also need to know the destination encoding. "ANSI" is not an encoding.
Source encoding could in theory be guessed based on the content, but specifying rules for such detection would be quite difficult. It would also likely be a highly error prone guess unless the input is long and happens to contain distinguishing special characters in typical arrangement. I hypothesise that a neural network could be trained to guess encodings. C++17 has no standard API for training or querying neural networks.

Unicode, UTF-8, UTF-16 and UTF-32 questions [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 8 years ago.
Improve this question
I read a lot about Unicode, ASCII, code pages, all the history, the invention of UTF-8, UTF-16 (UCS-2), UTF-32 (UCS-4) and who use them and so on, but I still having some questions that I tried hardly to find answers but I couldn't and I hope you to help me.
1 - Unicode is a standard for encoding characters and they specify a code point for each character. Something like U+0000 (example). Imagine that I have a file that has those code points (\u0000), in which point of my application I'm going to use it?
This might be a silly question but I really don't know in which point of my application I'm going to use it.
I'm creating an application that can read file that has those code points using the escape \u and I know that I can read it, decode it but now the next question.
2 - To which character set (code page) do I need to convert it? I saw some C++ libraries that they uses the name utf8_to_unicode or utf8-to-utf16 and also only utf8_decode, and this is what makes me confuse.
I don't know if will appear answers like this, but some might say: You need to convert it into code pages that you are going to use, but what if my application needs to be internationalized?
3 - I was wondering, in C++ if I try to display non-ASCII characters on terminal I got some confusing words. The question is: What makes the characters to be displayed are the fonts?
#include <iostream>
int main()
{
std::cout << "ö" << std::endl;
return 0;
}
The output (Windows):
├Â
4 - In which part of that process the encoding enter? It encodes, takes the code point and try to find the word that is equal on the fonts?
5 = WebKit is an engine for rendering web pages in web browsers, if you specify the charset as UTF-8 it works nicely with all characters, but if I specify another charset it doesn't, doesn't matter the font that I'm using, what happen?
<html>
<head>
<meta charset="iso-8859-1">
</head>
<body>
<p>ö</p>
</body>
</html>
The output:
ö
Works using:
<meta charset="utf-8">
6 - Imagine now that I read the file, I encode it, I have all the code points and I need to save the file again. Do I need to save it encoded (\u0000) or I need to decode first to transform again into characters and then save?
7 - Why the word "unicode" is a bit overloaded and is sometimes understood to mean utf-16? (source)
That's all for now. Thanks in advance.
I'm creating an application that can read file that has those code points using the escape \u and I know that I can read it, decode it but now the next question.
If you're writing a program that processes some kind of custom escapes, such as \uXXXX, it's entirely up to you when to convert these escapes into Unicode code points.
To which character set (code page) do I need to convert it?
That depends on what you want to do. If you're using some other library that requires a specific code page then it's up to you to convert data from one encoding into the encoding required by that library. If you don't have any hard requirements imposed by such third party libraries then there may be no reason to do any conversion.
I was wondering, in C++ if I try to display non-ASCII characters on terminal I got some confusing words.
This is because various layers of the technology stack use different encodings. From the sample output you give, "├Â" I can see that what's happening is that your compiler is encoding the string literal as UTF-8, but the console is using Windows codepage 850. Normally when there are encoding problems with the console you can fix them by setting the console output codepage to the correct value, unfortunately passing UTF-8 through std::cout currently has some unique problems. Using printf instead worked for me in VS2012:
#include <cstdio>
#include <Windows.h>
int main() {
SetConsoleOutputCP(CP_UTF8);
std::printf("%s\n", "ö");
}
Hopefully Microsoft fixes the C++ libraries if they haven't already done so in VS 14.
In which part of that process the encoding enter? It encodes, takes the code point and try to find the word that is equal on the fonts?
Bytes of data are meaningless unless you know the encoding. So the encoding matters in all parts of the process.
I don't understand the second question here.
if you specify the charset as UTF-8 it works nicely with all characters, but if I specify another charset it doesn't, doesn't matter the font that I'm using, what happen?
What's going on here is that when you write charset="iso-8859-1" you also have to actually convert the document to that encoding. You're not doing that and instead you're leaving the document as UTF-8 encoded.
As a little exercise, say I have a file that contains the following two bytes:
0xC3 0xB6
Using information on UTF-8 encoding and decoding, what codepoint do the bytes decode to?
Now using this 8859-1 codepage, what do the same bytes decode to?
As another exercise, save two copies of your HTML document, one using charset="iso-8859-1" and one with charset="utf-8". Now use a hex editor and examine the contents of both files.
Imagine now that I read the file, I encode it, I have all the code points and I need to save the file again. Do I need to save it encoded (\u0000) or I need to decode first to transform again into characters and then save?
This depends on the program that will need to read the file. If the program expects all non-ASCII characters to be escaped like that then you have to save the file that way. But escaping characters with \u is not a normal thing to do. I only see this done in a few places, such as JSON data and C++ source code.
Why the word "unicode" is a bit overloaded and is sometimes understood to mean utf-16?
Largely because Microsoft uses the term this way. They do so for historical reasons: When they added Unicode support they named all their options and setting "Unicode" but the only encoding they supported was UTF-16.

Understanding the concept of 'File encoding'

I have already gone through some stuff on the web and SOF explaining 'file encoding' but I still have questions. File is a group of related records and on disk, its contents are just stored as '1's and '0's. Every time, a running program wants to read in a file or write to the file, the file is brought into the RAM and put into the address space of the running program (aka process). Now what determines how the bits (or bytes) in the file should be decoded/encoded and read and displayed/written?
There is one explanation on SOF which reads 'At the storage level, a file contains an array of bytes. On top of this you have the encoding layer for text files. The format layer comes last, on top of the encoding layer for text files or on top of the array of bytes for all the other binary files'. I am sort of fine with this but would like to know if it is 100% correct.
The question basically came up when understanding file opening modes in C++.
I think the description of the orders of layers is confusing here. I would consider formats and encodings to be related but not tied together so tightly. Let's try to define it formally.
A file is a contiguous sequence of bytes. A byte is a contiguous sequence of bits.
A symbol is a unit of data. Bytes are one kind of symbol. There are other symbols that are not bytes. Consider the number 6 - it is a symbol but not a byte. It can however be encoded as a byte, commonly as 00000110 (this is the two's complement encoding of 6).
An encoding maps a set of symbols to another set of symbols. Most commonly, it maps from a set of non-byte symbols to bytes, which when applied to an entire file makes it a file encoding. Two's complement gives a representation of the numeric values. On the other hand, ASCII, for example, gives a representation of the Latin alphabet and related characters in bytes. If you take ASCII and apply it to a string of text, say "Hello, World!", you get a sequence of bytes. If you store this sequence of bytes as a file, you have a file encoded as ASCII.
A format describes a set of valid sequences of symbols. When applied to the bytes of a file, it is a file format. An example is the BMP file format for storing raster graphics. It specifies that there must be a few bytes at the beginning that identify the file format as BMP, followed by a few bytes to describe the size and depth of the image, and so on. An example of a format that is not a file format would be how we write decimal numbers in English. The basic format is a sequence of numerical characters followed by an optional decimal point with more numerical characters.
Text Files
A text file is a kind of file that has a very simple format. It's format is very simple because it has no structure. It immediately begins with some encoding of a character and ends with the encoding of the final character. There's usually no header or footer or metadata or anything like that. You just start interpreting the bytes as characters right from the beginning.
But how do you interpret the characters in the file? That's where the encoding comes in. If the file is encoded as ASCII, the byte 01000001 represents the Latin letter A. There are much more complicated encodings, such as UTF-8. In UTF-8, a character cannot necessarily be represented in a single byte. Some can, some can't. You determine the number of bytes to interpret as a character from the first few bits of the first byte.
When you open a file in your favourite text editor, how does it know how to interpret the bytes? Well that's an interesting problem. The text editor has to determine the encoding of the file. It can attempt to do this in many ways. Sometimes the file name gives a hint through its extension (.txt is likely to be at least ASCII compatible). Sometimes the first character of the file gives a good hint as to what the encoding is. Most text editors will, however, give you the option to specify which encoding to treat the file as.
A text file can have a format. Often the format is entirely independent of the encoding of the text. That is, the format doesn't describe the valid sequences of bytes at all. It instead describes the valid sequences of characters. For example, HTML is a format for text files for marking up documents. It describes the sequences of characters that determine the contents of a document (note: not the sequence of bytes). As an example, it says that the sequence of characters <html> are an opening tag and must be followed at some point by the closing tag </html>. Of course, the format is much more detailed than this.
Binary file
A binary file is a file with meaning determined by its file format. The file format describes the valid sequences of bytes within the file and the meaning that that sequence has. It is not some interpretation of the bytes that matters at the file format level - it is the order and arrangement of bytes.
As described above, the BMP file format gives a way of storing raster graphics. It says that the first two bytes must be 01000010 01001101, the next four bytes must give the size of the file as a count of the number of bytes, and so on, leading up to the actual pixel data.
A binary file can have encodings within it. To illustrate this, consider the previous example. I said that the four bytes following the first two in a BMP file give the size of the file in bytes. How are those bytes interpreted? The BMP file format states that those bytes give the size as an unsigned integer. This is the encoding of those bytes.
So when you browse the directories on your computer for a BMP file and open it, how does your system know how to open it? How does it know which program to use to view it? The format of a binary file is much more strongly hinted by the file extension than the encoding of a text file. If the filename has .bmp at the end, your system will likely consider it to be a BMP file and just open it in whatever graphics program you have. It may also look at the first few bytes and see what they suggest.
Summary
The first level of understanding the meaning of bytes in a file is that file's format. A text file has an extremely simple format - start at the beginning, interpreting characters until you reach the end. How you interpret the characters depends on that text file's character encoding. Most formats are more complicated, however, and will likely have encodings nested within them. At some level you have to start extracting abstract information from your bytes and that's where the encodings kick in. But then whatever is being encoded can also have a format that is applied to it. You have a chain of formats and encodings until you get the information that you want.
Let's see if this helps...
A Unix file is just an array of bits (1/0) the current minimum number of bits in a file is 8, i.e. 1 byte. All file interaction is done at no less than the byte level. On most systems now a days, you don't really have to concern your self with the maximum size of a file. There are still some small variences in Operating Systems, but very few if none have maximum sizes of less that 1 GB.
The encoding or format of a file is only dependent on the applications that use it.
There are many common file formats, such as 'unix ASCII text' and PDF. Most of the files you will come accross will have a documented format specification somewhere on the net. For example the specification of a 'Unix ASCII text file' is:
A collection of ascii characters where each line is terminated by a end of line character. The end of line character is specificed in c++ as std::endl' or the quoted "\n". Unix specifies this character as the binary value - 012(oct) or 00001010.
Hope this helps :)
The determination of how to encode/display something is entirely up to the designer of the program. Of course, there are standards for certain types of files - a PDF or JPG file has a standard format for its content. The definition of both PDF and JPG is quite complex.
Text files have at least somewhat of a standard - but how to interpret or use the contents of a text-file may be just as complex and confusing as JPEG - the only difference is that the content is (some sort of) text, so you can load it up in a text editor and try to make sense of it. But see below for an example line of "text in a database type application".
In C and C++, there is essentially just one distinction, files are either "binary" or "text" ("not-binary"). The difference is about the treatment of "special bits", mostly to do with "endings" - a text file will contain end of line markers, or newlines ('\n') [more in a bit about newlines], and in some operating systems , also contain "end of file marker(s)" - for example in old CP/M, the file was sized in blocks of 128 or 256 bytes. So if we had "Hello, World!\n" in a text file, that file would be 128 bytes long, and the remaining 114 bytes would be "end-of-file" markers. Most modern operating systems track filesize in bytes, so there's no need to have a end-of-file marker in the file. But C supports many operating systems, both new and old, so the language has an allowance for this. End of file is typically CTRL-Z (DOS, Windows, etc) or CTRL-D (Unix - Linux, etc). When the C runtime library hits the end of file character, it will stop reading and give the error code/behaviour, same as if "there is no more file to read here".
Line endings or newlines need special treatment because they are not always the same in the OS that the file is living on. For example, Windows and DOS uses "Carriage Return, Line Feed" (CR, LF - CTRL-M, CTRL-J, ASCII 13 and 10 respectively) as the end of line. In the various forms of Unix, (Linux, MacOS X and BSD for example), the line ending is "Line Feed" (LF, CTRL-J) alone. In older MacOS, the line ending is ONLY "carriage Return." So that you as a programmer don't have to worry about exactly how lines end, the C runtime library will do translation of the "native" line-ending to a standardized line-ending of '\n' (which translates to "Line Feed" or character value 10). Of course, this means that the C runtime library needs to know that "if there is a CR followed by LF, we should just give out an LF character."
For binary files, we really DO NOT want any translation of the data, just because our pixels happen to be the values 13 and 10 next to each other, doesn't mean we want it merged to a single 10 byte, right? And if the code reads a byte of the value 26 (CTRL-Z) or 4 (CTRL-D), we certainly don't want the input to stop there...
Now, if I have a database text file that contains:
10 01353-897617 14000 Mats
You probably have very little idea what that means - I mean you can probably figure out that "Mats" is my name - but it could also be those little cardboard things to go under glasses (aka "Beer-mats") or something to go on the floor, e.g. "Prayer Mats" for Muslims.
The number 10 could be a customer number, article number, "row number" or something like that. 01353-896617 could be just about anything - perhaps my telephone number [no it isn't, but it does indeed resemble it] - but it could also be a "manufacturers part number" or some form of serial number or some such. 14000? Price per item, number of units in stock, my salary [I hope not!], distance in miles from my address to Sydney in Australia [roughly, I think].
I'm sure someone else, not given anything else could come up with hundreds of other answers.
[The truth is that it's just made up nonsense for the purpose of this answer, except for the bit at the beginning of the "phone number", which is a valid UK area code - the point is to explain that "the meaning of a set of fields in a text-file can only be understood if there is something describing the meaning of the fields"]
Of course the same applies to binary files, except that it's often even harder to figure out what the content is, because of the lack of separators - if you didn't have spaces and dashes in the text above, it would be much harder to know what belongs where, right? There are typically no 'spaces' and other such things in a binary file. It's all down to someone's description or definition in some code somewhere, or something like that.
I hope my ramblings here have given you some idea.
Now what determines how the bits (or bytes) in the file should be decoded/encoded and read and displayed/written?
The format of the file, obviously. If you are reading a BMP file, you have to first read the header, then height*width pixel data. If you are reading .txt, just read the characters as-is. Text files can have different encodings, such as Unicode.
Some formats, like .png, are compressed, meaning that their raw data takes more space in memory that the file on disk.
The particular algorithm is chosen depending on various factors. On Windows, it's usually the extension that matters. In web, the content type is dominant.
In general, if you try to read the file in other format, you will usually get garbage. That can be forced sometimes : try opening a .bmp file in your text editor, for example.
So basically we're talking about text files mainly, right?
Now to the point: when your text editor loads the file into memory, from some information it deduces its file encoding (either you tell it or it has a special file format marker among the first few bytes of the file, or whatever). Then it's the program itself that decides how it treats the raw bytes.
For example, if you tell your text editor to open a file as ASCII, it will treat each byte as an individual character, and it will display the character A whenever encounters the number 65 as the current byte to show, etc (because 65 is the ASCII character code for A).
However, if you tell it to open your file as UTF-16, then it will grab two bytes (well, more precisely, two octets) at a time, it will use this so-called "word" as the numeric value to be looked up, and it will, for example, display a ç character when the two bytes it read were corresponding to 231, the Unicode character code of ç.

UTF 8 encoded Japanese string in XML

I am trying to create a SOAP call with Japanese string. The problem I faced is that when I encode this string to UTF8 encoded string, it has many control characters in it (e.g. 0x1B (Esc)). If I remove all such control characters to make it a valid SOAP call then the Japanese content appears as garbage on server side.
How can I create a valid SOAP request for Japanese characters? Any suggestion is highly appreciated.
I am using C++ with MS-DOM.
With Best Regards.
If I remember correctly it's true, the first 32 unicode code points are not allowed as characters in XML documents, even escaped with &#. Not sure whether they're allowed in HTML or not, but certainly the server thinks they're not allowed in your requests, and it gets the only meaningful vote.
I notice that your document claims to be encoded in iso-2022-jp, not utf-8. And indeed, the sequence of characters ESC $ B that appears in your document is valid iso-2022-jp. It indicates that the data is switching encodings (from ASCII to a 2-byte Japanese encoding called JIS X 0208-1983).
But somewhere in the process of constructing your request, something has seen that 0x1B byte and interpreted it as a character U+001B, not realising that it's intended as one byte in data that's already encoded in the document encoding. So, it has XML-escaped it as a "best effort", even though that's not valid XML.
Probably, whatever is serializing your XML document doesn't know that the encoding is supposed to be iso-2022-jp. I imagine it thinks it's supposed to be serializing the document as ASCII, ISO-Latin-1, or UTF-8, and the <meta> element means nothing to it (that's an HTML way of specifying the encoding anyway, it has no particular significance in XML). But I don't know MS-DOM, so I don't know how to correct that.
If you just remove the ESC characters from iso-2022-jp data, then you conceal the fact that the data has switched encodings, and so the decoder will continue to interpret all that 7nMK stuff as ASCII, when it's supposed to be interpreted as JIS X 0208-1983. Hence, garbage.
Something else strange -- the iso-2022-jp code to switch back to ASCII is ESC ( B, but I see |(B</font> in your data, when I'd expect the same thing to happen to the second ESC character as happened to the first: &#0x1B(B</font>. Similarly, $B#M#S(B and $BL#D+(B are mangled attempts to switch from ASCII to JIS X 0208-1983 and back, and again the ESC characters have just disappeared rather than being escaped.
I have no explanation for why some ESC characters have disappeared and one has been escaped, but it cannot be coincidence that what you generate looks almost, but not quite, like valid iso-2022-jp. I think iso-2022-jp is a 7 bit encoding, so part of the problem might be that you've taken iso-2022-jp data, and run it through a function that converts ISO-Latin-1 (or some other 8 bit encoding for which the lower half matches ASCII, for example any Windows code page) to UTF-8. If so, then this function leaves 7 bit data unchanged, it won't convert it to UTF-8. Then when interpreted as UTF-8, the data has ESC characters in it.
If you want to send the data as UTF-8, then first of all you need to actually convert it out of iso-2022-jp (to wide characters or to UTF-8, whichever your SOAP or XML library expects). Secondly you need to label it as UTF-8, not as iso-2022-jp. Finally you need to serialize the whole document as UTF-8, although as I've said you might already be doing that.
As pointed out by Steve Jessop, it looks like you have encoded the text as iso-2022-jp, not UTF-8. So the first thing to do is to check that and ensure that you have proper UTF-8.
If the problem still persists, consider encoding the text.
The simplest option is "hex encoding" where you just write the hex value of each byte as ASCII digits. e.g. the 0x1B byte becomes "1B", i.e. 0x31, 0x42.
If you want to be fancy you could use MIME or even UUENCODE.