Show errors when failed to read definitions file #201
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Makes sense to me |
gabelevi
added a commit
that referenced
this pull request
Dec 26, 2014
Show errors when failed to read definitions file
facebook-github-bot
pushed a commit
that referenced
this pull request
Oct 31, 2016
Summary:
The `hh_show` pseudo-function is very useful for debugging Hack type inference: it displays on the terminal the type that Hack believes an expression to have at the point that it type checks the line in which it occurs. (So, for example, it will show multiple times in a loop as Hack seeks a fixed point.)
But there is currently no way to dump any other aspect of Hack internals such as the type parameter environment. This diff rectifies this situation. It introduces a new nullary pseudo-function `hh_show_env` that causes Hack to dump information about its environment, including:
* The types of all locals (`local_types`)
* The lower and upper bounds on all type parameters in scope (`tpenv`)
* The state of the type-variable-to-type variable substitution (`subst`)
* The state of the type-variable-to-type environment (`tenv`)
The last two aren't shown in full but instead as a delta from the previous time that `hh_show_env` was called.
Some other aspects:
* Colour display, with indentation!
* The position information of the `hh_show_env` command is shown above the environment information, with an iteration number in square brackets (because it might be invoked more than once e.g. round a loop).
The `hh_show` command is also improved, to display this position information.
Finally, type variable numbers are shown in types so that the `subst` and `tpenv` parts of the environment can be debugged.
Here is an example: the test `erling_loop5.php` with `hh_show_env()` inserted at the end of the inner loop. The output from `hh_single_type_check` is as follows:
```
File "erling_loop5.php", line 34, characters 5-17:[1]
local_types
$obj: X as X<#216(int | string)>
$x: (int)
$y: (int)
$z: (int | string)
$i: int
subst(changes)
#199: #216; #201: #212; #212: #214; #214: #216;
tenv(changes)
#198: [unresolved]; #199: [unresolved]; #200: [unresolved]; #201: int; #212: int;
#214: int; #216: (int | string);
File "erling_loop5.php", line 34, characters 5-17:[2]
local_types
$obj: X as X<#223(int | string)>
$x: (string | int)
$y: (int | string)
$z: (int | string)
$i: int
subst(changes)
#223: #223; #216: #219; #219: #221; #221: #223;
tenv(changes)
#219: (string | int); #221: (int | string); #223: (int | string);
```
Reviewed By: dabek
Differential Revision: D4068570
fbshipit-source-id: cb96d5178bc41f73a8ddd8776ab7f91f69de2b71
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When I make a typo in a huge declarations file, I get common error
Can't read library definitions file %s, exiting.Is there any special reason not to show the exact errors while parsing definitions files? Since
--no-flowlibis now available (and many thanks to merge) the question becomes relevant.