You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The `bufmode_t` enum was added to `src/support/ios.h` in
d316c93 by Jeff Bezanson with the
comment
```c
// this flag controls when data actually moves out to the underlying I/O
// channel. memory streams are a special case of this where the data
// never moves out.
typedef enum { bm_none, bm_line, bm_block, bm_mem } bufmode_t;
```
In 30aed86 Keno Fischer added a new
comment and modified the enum:
```diff
// this flag controls when data actually moves out to the underlying I/O
// channel. memory streams are a special case of this where the data
// never moves out.
-typedef enum { bm_none, bm_line, bm_block, bm_mem } bufmode_t;
+
+//make it compatible with UV Handles
+typedef enum { bm_none=UV_HANDLE_TYPE_MAX+1, bm_line, bm_block, bm_mem } bufmode_t;
```
In c12aca8 Jameson Nash removed `uv.h`
and updated the enum definition to no longer use `UV_HANDLE_TYPE_MAX`:
```diff
-typedef enum { bm_none=UV_HANDLE_TYPE_MAX+1, bm_line, bm_block, bm_mem } bufmode_t;
+typedef enum { bm_none=1000, bm_line, bm_block, bm_mem } bufmode_t;
```
This last change made the comment by Keno Fischer seemingly obsolete,
yet the comment remains. This pull request simply removes the two lines
remaining from Keno Fischer's commit. Perhaps it is better to update the
comment to explain why `bm_none` now has the value `1000` but I did not
dig deep enough into the changes to figure out if this value is
significant in some way that should be documented.
0 commit comments