I am using winapi's CreateFile() and WriteFile() functions to open the file in append mode and write into the file.
When i use FILE_APPEND_DATA flag, and then write into the file, it append the new item at the end of the file, but i want to append the new data at the beginning of the file.
I have been through many forums and the official documentation of the winapi but not able to find the solution can anyone help me with it.
Just because the you wish that "append" were to mean "insert data at any random position" doesn't make it do that. "Appending" has an unambiguous meaning, and opening a file in FILE_APPEND_DATA mode provides just that.
There is no system-supported way to open a file in "prepend" mode (which is apparently what you're looking for). You'll have to find another solution to your problem, or a better problem to solve.
Related
I have to read from a file, do some operations based on the data I got from that file and then rewrite the whole file with new values obtained after the operations were made. I tried
fstream file("date.in", ios::in|ios::out)
but seems like it puts the new set of data at the end of file. Also tried
fstream file("date.in", ios::in|ios::out|ios::trunc)
but then I can't even read the first set of data as it appears not to be there.
If you want to read and write to the same offsets in the file, you can set the put and get pointers with seekp() and seekg(). Their documentation can be found at cppreference.com.
I'm trying to create a reference program which I think will use an excel spreadsheet to hold information for reading only. I want the user to be able to select a topic from an option list and have the information in the appropriate cell be fed back to them. The program is being written in C++. My question is, how do I access specific cells from a spreadsheet from my program? I've researched it a little and I've seen that I want to save my file as a csv and use fscanf to read the contents, but I'm at a loss as to how I would do this part. I googled it and found this thread:
http://www.daniweb.com/software-development/cpp/threads/204808/parsing-a-csv-file-separated-by-semicolons
but I think it reads in all of the data from the CSV? From what I can tell anyways. And I only want to pull specific elements. Is that possible?
If you only want specific elements, you would still have to parse all contents of the file until you reach those elements. You don't have to store values you don't need, but you do need to parse them to advance in the file.
Are you invoking the program from Excel? If you are, a little VBA goes a long way. You could always only export the cells of interest ready for your C++ program to read in.
Otherwise, other answers are correct. However, you don't need to load the entire file into memory at once. You can use std::fstream to open the file and read in each line of the file, parsing in the required information for each line.
I have a large text file.
Each time my program runs, it needs to read in the first line, remove it, and put that data back into the bottom of the file.
Is there a way to accomplish this task without having to read in every part of the file?
It would be great to follow this example of pseudo code:
1. Open file stream for reading/writing
2. data = first line of file
3. remove first line from file <-- can I do this?
4. Close file stream
5. Open file stream for appending
6. write data to file
7. Close file stream
The reason I'm trying to avoid reading everything in is because the program runs at a specific time each day. I don't want the delay to be longer each time the file gets bigger.
All the solutions I've found require that the program process the whole file. If C++ filestreams are not able to accomplish this, I'm up for whatever alternative is quick and efficient for my C++ program to execute.
thanks.
The unfortunate truth is that no filesystem on a modern OS is designed to do this. The only way to remove something from the beginning of a file is to copy the contents to a new file, except for the first bit. There's simply no way to do precisely what you want to do.
But hopefully you can do a bit of redesign. Maybe each entry could be a record in a database -- then the reordering can be done very efficiently. Or perhaps the file could contain fixed-size records, and you could use a second file of indexes to specify record order, so that rearranging the file was just a matter of updating the indices.
I would like my Browser Helper Object which is simply a .dll, to be able to read a text file. I have tried using a pointer to a FILE, as well as ifstream in("file name goes here"). Before implementing these two methods of reading files into the BHO, I tested them individually, and made sure each example dealt with similar data types and locations. Both of them worked without a problem, yet testing the BHO reveals that the file cant even be opened. I have searched google for an alternative method, and after exhausting all other options, I'm hoping that someone would be able to provide me with some guidance/resource. Anything is appreciated; I will keep trying to find a solution and will post what I can in the event that someone else may have the same problem.
Are you providing an absolute path to the file? If you're just using a relative path you may not be in the same working directory while running IE.
I think you might be a victim of the IE protected mode.
http://www.codeproject.com/KB/vista-security/PMSurvivalGuide.aspx
Under protected mode your addin might not have access to registry and file system as you might like.
Apparently this supposed to be possible. For example opening and operating on a file with NOTEPAD, or HxD. But aren't they all text files...how would one specify which text editor to open the file and operate on the file with using the WINDOWS API. It is certainly not in "CreateFile".
Hopefully I'm understanding your question... The easiest way to do this is to launch the desired editor and pass the filename as an argument, rather than "invoking" the file (which will launch the default program associated with the file type).
For example, notepad.exe mytextfile.txt or gvim.exe mytextfile.txt.
If the editor is not on your %PATH%, you'll need to use a full path file name.
What are you trying to do, exactly? You could:
Maintain a list of editors that you expect to be installed and have entries for in the system's PATH (bad idea)
Have an editor/editors that you want to use, query the Windows registry to find the installation path of the editors (using RegGetValue), and launch the editor with CreateProcess) (a little better idea)
Query the registry to get the default editor for a given file type and then launch that editor using CreateProcess. (best idea)
But it all depends on what your goal is really.
Edit based on requirements
So, just so we're on the same page, from C++, you want to:
Take a command line parameter to your C++ application (filename)
Open that file in an arbitrary editor
Detect when the user has made changes to that file
Operate on the file contents
Is that correct?
If so, you could:
Use Boost libs to compute a CRC for the current data in the file
Launch an editor using one of the methods I initially described
Stick in a tight loop and sleep so you don't chew up resources while the initially computed CRC matches one calculated every iteration of the loop
Of course, there are all kinds of issues that you'd have to deal with (that's just a super simple way of describing the algorithm I might use), such as:
What happens if the user doesn't change the file?
What happens if the file isn't found?
I'm sure that there are a number of different methods of doing this, but this is the easiest method that I can think of at the moment (while still being able to be fairly certain of the changes).
Disclaimer: I haven't implemented something like this, so I might be completely off base ;)
Are you looking for the ShellExecute() or ShellExecuteEx() APIs on Windows? They'll launch whatever program is registered for a file (generally based on the filename extention).