C++Talk.NET C++ language newsgroups   Archives   FAQ   Search   Memberlist   Usergroups   Register   Profile   Log in to check your private messages   Log in  why does ofstream(..., ate) file          C++Talk.NET Forum Index -> C++ Language (Moderated) View previous topic :: View next topic   Author Message lotusny78@yahoo.com Guest Posted: Wed Oct 12, 2005 12:58 am    Post subject: why does ofstream(..., ate) file When using VC++ .Net, the statement: ofstream myFile(orderFileName.c_str(), ios_base::ate); causes the file to be truncated. 1) Is this a bug or am I going crazy? 2) What's the point of seeking to the end of the file if you're just going to truncate it? 3) Also, if you open with ios_base::app, can you move the put pointer to the middle of the file and write there later on? [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top kanze Guest Posted: Wed Oct 12, 2005 8:50 am    Post subject: Re: why does ofstream(..., ate) file [email]lotusny78 (AT) yahoo (DOT) com[/email] wrote: Quote: When using VC++ .Net, the statement: ofstream myFile(orderFileName.c_str(), ios_base::ate); causes the file to be truncated. 1) Is this a bug or am I going crazy? Your just mixing issues:-). The ios::ate flag has no effect on how the file is initially opened. All it does is request a seek to the end immediately after opening. Quote: 2) What's the point of seeking to the end of the file if you're just going to truncate it? Good question. ios::ate is pretty useless when opening a file for writing. It's also pretty useless when opening a file for reading, since it means that the first read will automatically return end of file. And if you're doing both reading and writing, you'll be doing your own seeking anyway, so what's the point? Quote: 3) Also, if you open with ios_base::app, can you move the put pointer to the middle of the file and write there later on? No. The point of ios::app is precisely that every write will be to the end of the file. The intent is for this to be atomic, if the system supports it. Something you can't do otherwise. I'm not sure what you're trying to do. As Plauger recently pointed out in response to another question (involving mixing peek() and seek() on the same stream), iostreams supports two models: a streams model, and a random access, and the two don't work well together. (My experience has been that the random access model is sort of a hack anyway, and only works on binary streams.) With regards to open: ios::app applies to the streaming model, and only when writing; one of the goals is to allow correct streaming when several different processes are writing to the same file. (Whether it is really appropriate with the iostream formatting model is another question. It really only works if you format your data into a string, and output that in a single operation.) Personally, I don't quite see what the use of ios::ate is. But the semantics are defined in terms of seek, so I would presume that it is only relevant for the random access mode. And as far as I can tell, the random access mode only really works well files are opened for bidirectional access (read and write), and in binary mode. (There are some things you can do in text mode, but they are limited by the fact that in text mode, the parameter to seek must be a value previously returned by tell, unless it is to the beginning or the end of the file.) So if random access is desired, you should open the file with: ofstream myFile( orderFileName.c_str(), ios::in | ios::out | ios::binary ) ; (Strictly speaking, the ios::out isn't necessary, since the open function in ofstream will automatically or it in. I just find it a bit clearer -- otherwise, the presence of ios::in suggests only input.) If you do this, you should be able to add ios::ate to the flags as well, without any problem. Or you could just seek to the end immediately after the open. Note that some systems do guarantee more than the standard; under Unix, for example, ios::binary is actually a no-op, and isn't necessary for seeking. Under Windows, if you know where the newlines are in the file, and count each newline for two bytes, you can also seek to arbitrary positions in text files. But these are system specific extensions, and not guaranteed by the standard. -- James Kanze GABI Software Conseils en informatique orientée objet/ Beratung in objektorientierter Datenverarbeitung 9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34 [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top anas.hashmi@gmail.com Guest Posted: Sun Oct 16, 2005 9:54 am    Post subject: Re: why does ofstream(..., ate) file I am trying to write to the beginning of a file. The reason: I want to make a form where webmasters can use it to insert in updates to a webpage without having to go directly into the file and start editing. The form must write to the beginning of file so that the soonest update appears first. I am actually writing in PHP but see that they do not support it. So I turned to C++. On About.com, the tutorial on C++ Lesson 12 says that ios::ate is used to write anywhere in the file without truncating. So i did a search on how I can manipulate that to the beginning of a file. I could not find anything. Is there anyone who can help solve this problem? [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Valentin Samko Guest Posted: Sun Oct 16, 2005 3:49 pm    Post subject: Re: why does ofstream(..., ate) file [email]anas.hashmi (AT) gmail (DOT) com[/email] wrote: Quote: I am trying to write to the beginning of a file. The reason: I want to make a form where webmasters can use it to insert in updates to a webpage without having to go directly into the file and start editing. The form must write to the beginning of file so that the soonest update appears first. I am actually writing in PHP but see that they do not support it. So I turned to C++. On About.com, the tutorial on C++ Lesson 12 says that ios::ate is used to write anywhere in the file without truncating. So i did a search on how I can manipulate that to the beginning of a file. I could not find anything. Is there anyone who can help solve this problem? You can not write to the beginning of a file without overwriting existing data, i.e. you can not "insert" data anywhere but at the end. This has nothing to do with C++ or PHP. This is how file systems work. -- Valentin Samko - http://val.samko.info [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top James Kanze Guest Posted: Sun Oct 16, 2005 4:16 pm    Post subject: Re: why does ofstream(..., ate) file [email]anas.hashmi (AT) gmail (DOT) com[/email] wrote: Quote: I am trying to write to the beginning of a file. The reason: I want to make a form where webmasters can use it to insert in updates to a webpage without having to go directly into the file and start editing. The form must write to the beginning of file so that the soonest update appears first. I am actually writing in PHP but see that they do not support it. So I turned to C++. On About.com, the tutorial on C++ Lesson 12 says that ios::ate is used to write anywhere in the file without truncating. So i did a search on how I can manipulate that to the beginning of a file. I could not find anything. It sounds to me like you are trying to insert at the beginning of the file. This isn't possible in C++, for the simple reason that it isn't possible under many (most?) operating systems; if you position at the beginning of the file and write, you will overwrite what was previously there. The only way to work around this restriction is to copy the file. Something along the lines of: create temporary file write new data to temporary file copy existing file to temporary file delete existing file rename temporary file I won't go into the details here, because they shouldn't be a problem for an experienced C++ programmer, and because you say you were working in PHP, and I would be very surprised if you cannot do the same thing in PHP. Since the solutions do abandon C++ entirely, I might as well go all the way: if you store the Web page data as SQL records, it should be possible to insert the new data without copying the old. Just associate a timestamp with each blob, and retrieve sorted by timestamps. -- James Kanze mailto: [email]james.kanze (AT) free (DOT) fr[/email] Conseils en informatique orientée objet/ Beratung in objektorientierter Datenverarbeitung 9 pl. Pierre Sémard, 78210 St.-Cyr-l'École, France +33 (0)1 30 23 00 34 [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top anas.hashmi@gmail.com Guest Posted: Tue Oct 18, 2005 10:28 am    Post subject: Re: why does ofstream(..., ate) file Thanks. create temporary file write new data to temporary file copy existing file to temporary file delete existing file rename temporary file the suggestion here is what i ended up doing instead. or kind of. at first, i thought by deleting existing file, i would make the file unavailable... but all this happens in a matter of nanoseconds so it was not a problem. Thanks a whole lot everyone. [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top kanze Guest Posted: Wed Oct 19, 2005 8:50 am    Post subject: Re: why does ofstream(..., ate) file [email]anas.hashmi (AT) gmail (DOT) com[/email] wrote: Quote: Thanks. create temporary file write new data to temporary file copy existing file to temporary file delete existing file rename temporary file the suggestion here is what i ended up doing instead. or kind of. at first, i thought by deleting existing file, i would make the file unavailable... but all this happens in a matter of nanoseconds so it was not a problem. It could be a problem if your system crashed precisely in those nanoseconds. It's a fairly common practice when using this technique to use recognizable names for the temporary files, and for the application to look for them on start-up, and initiate recovery procedures. An important point concerning the above sequence is that there is never a moment when neither the new nor the old version are on disk, so there is never a moment when you cannot recover one of the two. (Note that this is also an argument against in place modification, even when possible. With in place modification, there may be times when the only copy on disk is in an incoherent state.) -- James Kanze GABI Software Conseils en informatique orientée objet/ Beratung in objektorientierter Datenverarbeitung 9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34 [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Display posts from previous: All Posts1 Day7 Days2 Weeks1 Month3 Months6 Months1 Year Oldest FirstNewest First         C++Talk.NET Forum Index -> C++ Language (Moderated) All times are GMT Page 1 of 1   Jump to: Select a forum C/C++----------------C LanguageC Language (Moderated)C++ Language (Moderated)C++ language (comp.lang.c++)C++ language, library and standards Non-English C/C++ Newsgroups----------------C++ (German)C++ (French)  You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum Powered by phpBB © 2001, 2006 phpBB Group SEO toolkit © 2004-2006 webmedic.