I have a project I am working on in C++ using VC++ 6 and I am reading in a text file that is output from a batch that runs a series of dir/b *.extension>>list.txt routines to append to file all file extensions I want to have in my list file instead of just all files in that directory. This is used to list all the movie/VIDEOS to this text file. The C++ is used to create/compile a HTML page for my HTPC.

Currently I am able to open the list.txt file and add each line to the creation of an HTML page that places that lines contents as a hyperlink in an alphabetical listing, but instead of listing it as say movie1.mp4, I want to list it as movie1. So I was thinking that if I trimmed 4 characters off of the line, that would solve it, but for some files with 4 character file extensions instead of 3 character file extensions I would then have to trim 5 characters to include the .mp4 or .mpeg to remove the dot as well.

So I was thinking that maybe I can read in the line, then run a process that looks for the dot "." and it doesnt matter then how short or long a file extension is. Only problem is I have never done this before and so I have stumped myself. Did some searching in google and figured I'd post here since I didnt find anything close to what I am trying to accomplish.

In the end the C++ program writes/compiles a htpc.htm webpage that I have linked to the startup of the htpc, so I can pick movies and music etc from an organized menu. I was editing this htpc.htm page manually and thought that I should really just write a program that will import file names and then link them in a html webpage vs having to go in and add the additional line into my html page each time I add new videos and music etc.Why have the batch file to begin with?Hello BC... well I have the batch to begin with mainly because I am using the dir/b to get directory listing. I could have snuck that into the C++ with a system("dir/b *.mp4>>list.txt"); type call but been staying away from using system calls as for I read that they are a bad habit to use in C++. So I figured, I'd run a quick batch, then have my list.txt file ready for me with the files I want to have included in the created HTPC webpage and then use the C++ to then read in this text file and run with it.

*If you can show me a technique to output a file list to a file avoiding the system call in C++, I would be very interested in seeing how this is done. I am sure it can probably be done, but this is new territory accessing a file system to then pass that information (filename.extension) listing to a file vs using DOS commands.

 I have posted a very basic draft of what I have. This isnt the web page I am currently using, just a quick proof of concept sort of test, which generates the HTPC.htm page. HTML is easy to make pretty later on with frames and tables etc later on. I also need to add code to delete the HTPC.htm file if it exists prior to writing to it -or- clear the file first to avoid a double appending of HTML code added to the end of the initial. Deleting it I would have to use a system call, but I might be able to wipe its contents using C++ that I will have to figure out how to pass nothing to it to overwrite something with nothing..lol  then be able to write cleanly to it. I am guessing that i can write overtop of it with a single space " " probably since if it appends the after a single space, the webpage will still act the same as the browser translates it.

Being able to trim off .mp4 and .mpeg from the name of the file listed in the html hyperlink displayed text would be neat to do to clean this up I figured since I just want titles listed and the file entension showing is kind of ugly. I figured there must be a way to find "." and trim "." and the next remaining characters no matter how many there are on that line without affecting the next line below it. But still stumped


Here is test contents of link.txt
Code: [Select]file1.mp4
file2.mp4
file3.mp4
file4.mp4
file5.mp4



And here is my current code, which may look a little ugly without comments and I merged a book reference and online reference of code for technique and then tweaked them together to get what I needed since I am not good at just coding at this level without looking in my books or online etc unless I want to battle it out when compiling and fixing the issues.

Code: [Select]#include<iostream>

using std::cerr;
using std::cout;
using std::endl;

#include<fstream>
using namespace std;
using std::ifstream;

#include<cstdlib>


int main()
{
ifstream indata;
char buffer1[2048];
indata.open("link.txt");
if(!indata){
cerr<<"Error: file could not be opened"<<endl;
exit(1);
}
indata>>buffer1;

ofstream myfile1;
myfile1.open ("htpc.htm",ios::out|ios::app);

myfile1<<"<html>"<<endl;
myfile1<<"<body>"<<endl;
myfile1<<"<center>"<<endl;
myfile1.close();
while(!indata.eof()){
ofstream myfile2;
myfile2.open ("htpc.htm",ios::out|ios::app);

myfile2<<"<a href\""<<buffer1<<"\">"<<buffer1<<"</a><BR>"<<endl;

indata>>buffer1;
myfile2.close();
}
ofstream myfile3;
myfile3.open ("htpc.htm",ios::out|ios::app);
myfile3<<"</body>"<<endl;
myfile3<<"</html>"<<endl;

myfile3.close();




indata.close();
cout<<"End-of-file reached.."<<endl;
return(0);
} Code: [Select]#include <windows.h>
#include <iostream>
#include <fstream>
#include <string>
using namespace std;
//TASK: enumerates all MP3 files in a given directory and creates a simple HTML file with links for each one.
int _tmain(int argc, _TCHAR* argv[])
{
WIN32_FIND_DATA* fdata = (WIN32_FIND_DATA*)malloc(sizeof(WIN32_FIND_DATA));

//Change this as needed to use a different folder.
//change this to change the folder. Possibly make command line argument, who knows.
//make sure to change the HREF line below if you change this, too.
HANDLE fhandle = FindFirstFile(L"D:\\music\\*.mp3",fdata);
BOOL filesfound=TRUE;
if(fhandle==INVALID_HANDLE_VALUE)
{
cout << GetLastError();
filesfound=FALSE;

}
ofstream htmlout;
//delete the file if it exists....
DeleteFile(L"D:\\htpc.htm");
htmlout.open ("D:\\htpc.htm",ios::out|ios::app);
htmlout<< "<html><body><center>";
if(!filesfound) htmlout << "<h2>No Files.</h2>";
if(fhandle!=INVALID_HANDLE_VALUE)
{
for(;;)
{
//assumes fdata->cFileName is a WCHAR_T.
char buff[MAX_PATH];
//wcstombs ( char * mbstr, const wchar_t * wcstr, size_t max );
ZeroMemory(buff,MAX_PATH);
wcstombs(buff,fdata->cFileName,MAX_PATH);
string outputname;
outputname.assign(buff);
cout << "found file:" ;
cout << buff << endl;

int dotpos = outputname.rfind(".");
string titlename = outputname.substr(0,dotpos);

cout << "name:" << titlename;
htmlout << "<a href=\"" << "D:\\music\\" << outputname << "\">" << titlename << " </a><br>\n";
ZeroMemory(fdata,sizeof(WIN32_FIND_DATA));
BOOL fnf=FindNextFile(fhandle,fdata);
if(fnf==0) break;
}
}
htmlout << "</center></body></html>";
FindClose(fhandle);
htmlout.close();
//int test;
//cin>>test;

free(fdata);
return 0;
}


Dave, why don't you just make the original batch file strip off the extensions? And build the html page too? Isn't writing a C++ program to do this using a sledgehammer to crack a nut?Thanks BC for posting the code, I am going to copy/paste and compile and look through it to learn how it works. And may post questions later. Really interested in how it addresses the file system to retreive the info that the dir/b does in that batch.

Hello Salmon.... well the situation I have is that I know some of both batch and C++ with C++ being my stronger skill when dealing with reading/writing from and to files and dealing with variables and strings and am a master of neither, so I end up making lots of ugly mixed program/scripts to get the job done usually.

I guess I went the route of C++ also because I know that batch has limitations and C++ can do all. I mainly didnt know how to perform a process similar to dir/b to access a file system to retrieve file names and figured the batch was the quick band aid to create the information that I can then feed to the C++ program.

*If you have a moment to show how this can be done in batch that would be great. As for I am interested in learning how it can be done by example. It would be neat to see how to pass the contents of dir/b into variables or read in the listing from dir/b that was written to list.txt at the start of the batch to then output that to an html file as well as trimming off the file extensions. I am guessing its mainly just creation of the html file initially by a bunch of Quote

echo. <HTML><BODY>>>htpc.htm

echo. <a href"(variable [A] here to pass full filename and extension)">(variable here to pass the full file name without extension) [/url]>>htpc.htm

I came up with this but it's useless and doesn't work for some reason.

--------------------Configuration: htpc3 - Win32 Debug--------------------
Compiling...
htpc3.cpp
c:\dchtml\htpc3.cpp(7) : error C2061: syntax error : identifier '_TCHAR'
c:\dchtml\htpc3.cpp(14) : error C2664: 'FindFirstFileA' : cannot convert parameter 1 from 'unsigned short [15]' to 'const char *'
        Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast
c:\dchtml\htpc3.cpp(24) : error C2664: 'DeleteFileA' : cannot convert parameter 1 from 'unsigned short [12]' to 'const char *'
        Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast
c:\dchtml\htpc3.cpp(36) : error C2664: 'wcstombs' : cannot convert parameter 2 from 'char [260]' to 'const unsigned short *'
        Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast
Error executing cl.exe.

htpc3.obj - 4 error(s), 0 warning(s)

The other thing I need to do is stay active with it too instead of just coding when I need something, since if you dont use it you lose it.

Only complaint I have with them is that whatever you write you need to make sure that Dot NET 3.5 I think the version is is installed on any system that you execute the compiled programs from. Its too bad they cant squeeze that into the .EXE so it can be stand alone I guess you could call it vs requiring the 3.5 framework that most systems I have seen dont normally update to with MS updates. I think the MS updates bring systems to Dot NET 2.0.

I also have to change SET searchpath=%cd% to SET searchpath=%cc% , which was a guess on my part

So I figured, I'd run a quick batch, then have my list.txt file ready for me with the files I want to have included in the created HTPC webpage and then use the C++ to then read in this text file and run with it.