Various bugfixes and improvements around buffer md5 hash calculation and fastdirty
#3430
Conversation
According to the Go hash package documentation [1]:
type Hash interface {
// Write (via the embedded io.Writer interface) adds more data to the running hash.
// It never returns an error.
io.Writer
[1] https://pkg.go.dev/hash#Hash
Make calcHash() respect the buffer's file endings (unix vs dos), to make its calculation of the file size consistent with how we calculate it in other cases (i.e. when opening or saving the file) and with the `fastdirty` option documentation, i.e. make calcHash() return ErrFileTooLarge if and only if the exact file size exceeds 50KB.
This behavior is then aligned to the actual documentation of `fastdirty`. Additionally set the origHash to zero in case the buffer was already modified.
When we have already enabled `fastdirty` but have not updated origHash yet, we shouldn't use Modified() since it depends on origHash which is still outdated, and thus returns wrong values. This fixes the following issue: enable `fastdirty`, modify the buffer, save the buffer and disable `fastdirty` -> micro wrongly reports the buffer as modified (whereas it has just been saved). Note that this fix, though, also causes a regression: e.g. if we run `set fastdirty false` while fastdirty is already disabled, micro may unexpectedly report a non-modified buffer as modified (in the case if isModified is true but the buffer it actually not modified, since its md5 sum matches and fastdirty is disabled), since this fix assumes that since we are disabling fastdirty, it has been enabled. This shall be fixed by PR zyedidia#3343 which makes `set` do nothing if the option value doesn't change.
Let calcHash() unconditionally hash whatever buffer it is asked to hash, and let its callers explicitly check if the buffer is too large before calling calcHash(). This makes things simpler and less error-prone (no extra source of truth about whether the file is too large, we don't need to remember to check if calcHash() fails, we can be sure calcHash() will actually update the provided hash), and actually faster (since just calculating the buffer size, i.e. adding line lengths, is faster than md5 calculation). In particular, this fixes the following bugs: 1. Since ReOpen() doesn't check calcHash() return value, if the reloaded file is too large while the old version of the file is not, calcHash() returns ErrFileTooLarge and doesn't update origHash, so so Modified() returns true since the reloaded file's md5 sum doesn't match the old origHash, so micro wrongly reports the newly reloaded file as modified. 2. Since Modified() doesn't check calcHash() return value, Modified() may return false positives or false negatives if the buffer has *just* become too large so calcHash() returns ErrFileTooLarge and doesn't update `buff`.
Similarly to how we force `fastdirty` to true when opening a large file (when creating the buffer), force it also when reopening a file, in case the file on disk became large since we opened it.
dmaluka
force-pushed
the
calchash-fixes
branch
from
57edb8f
to
0b15b57
Compare
dmaluka
changed the title
fastdirtyfastdirty
|
Shouldn't we check the actual buffer size in |
|
I was thinking about that, but it's not so non-controversial. For example, if the user accidentally pastes a large chunk of text so that the file grows above the limit, we enable ...Honestly, this md5 hashing feature always seemed dubious to me. I think it would be better if we just had better support for basic detection: return non-modified status not only if the buffer was not modified but also if all its modifications have been undone, i.e. if the undo stack is empty (or with This entire md5 feature seems like a workaround for the lack of this basic functionality. And it consumes much more CPU, of course. And I'm actually planning to implement this basic functionality, for better detection when |
I was thinking about only returning |
|
Hmm, maybe... OTOH that would be an undocumented behavior, and not consistent with other cases (in other cases we override |
|
Ok, acceptable. |
Or at least disable it by default (I suppose most users would be fine with "unmodified == all modifications undone" instead of "unmodified == file content not changed", and micro will be more CPU thrifty). |
See commit messages for details.