C ++ fstream Removing file contents from selected point

Question

I need to delete the contents of a file from a selected point (C ++ fstream), which function should I use?

I wrote objects , I need to delete these objects in the middle of the file

Answer 1

C ++ does not have a standard mechanism for trimming a file at a given point. You either need to recreate the file (open with ios::trunc and write the content you want to save), or use the API calls for the OS ( SetEndOfFile on Windows, truncate or ftruncate on Unix).

EDIT: Removing material in the middle of a file is extremely dangerous. Long before considering any other alternatives, I would try to use a server-side database engine, such as SQLite, to store serialized objects. Even better, I would use SQLite as intended by storing the data needed by these objects in the correct schema.

EDIT 2: If the problem instruction requires access to the unprocessed file ...

As a rule, you do not delete data from the middle of the file. If the objects can be serialized to a fixed size on the disk, you can work with them as records, and instead of trying to delete the data, you use a table that indexes the records in the file. For example, if you record four records in sequence, the table will contain [0, 1, 2, 3] . To delete the second entry, you simply delete its entry from the table: [0, 2, 3] . There are at least two ways to reuse the holes left by the table:

• In each insert, scan the first unused index and write the object to the appropriate recording location. It will be more expensive though, as the file grows.
• Keep a free list. Store the index of the most recently released record as a separate variable. In the space occupied by this record, the index of the freed record is encoded, etc. This maintains a handy list of free entries, but it only requires one extra number. However, it’s more difficult to work with, and an additional I / O disk is required during removal and insertion.

If the objects cannot be serialized to a fixed length, then this approach becomes much more complicated. The variable length write control code is very complex.

Finally, if a task operator requires saving records to disk on disk, then this is a stupid problem, because inserting / deleting in the middle of a file is ridiculously expensive; no reasonable design would require this.

Answer 2

The general method is to open the file for read access, open a new file for write access, read the contents of the first file and write the data that you want to save to the second file. Upon completion, you delete the first file and rename the second to the first.