Because as @hwillson rightfully points out:
> that's what we're really exposing and `connect` themselves refer to the container that holds the defined middleware (the result of calling the `connect()` function) as a connect "app"."
As per the comment on line 653, this appears to be the best (only?) way to let users do things like set up a custom error page to catch app-rendering-time exceptions.
Keeping track of the latest version for each module identifier was more
important when we were using localStorage, because it helped with
periodically removing unused module sources.
Since the storage limits for IndexedDB are substantially higher than for
localStorage, we can safely keep old versions in the cache, and reap the
benefits if those versions are ever needed again.
Eliminating this extra logic simplifies the caching logic considerably,
and also improves cache read performance.
This reverts commit 027d3e15a7.
This optimization was not safe in all browsers. In particular, Firefox
complained about requests against inactive transactions.
If import(...) is called multiple times in quick succession, these changes
should allow cache.checkMany to avoid the overhead of creating multiple
IndexedDB transactions.
This prevents __dynamicImport from blocking other method calls made by the
application, but introduces the possibility that __dynamicImport method
results could be delivered out of order, which is now handled in the
fetchMissing function.
I was able to achieve tolerable performance by using only one (read-only)
transaction for each cache.getMany call, and only one (read+write)
transaction for each cache.setMany call.
Apply the check for a SyntaxError indicating an incomplete command only
to the ECMAScript translation and parsing of a command and not to the
execution, following the example of Node's defaultEval function.
Fixes#8290.
This restriction was preventing the standard-minifier-css from getting
the latest updates to minifier-css. Typically, when publishing
a package outside of a release cycle, I believe we need to manually add
this constraint since `meteor publish-release` normally takes care of it
automatically, however in most cases, I don't believe this should be
(permanently) committed.
I may stand to be corrected, but will find out soon!
This was only intended to be there during the publishing of
`standard-minifier-js` as part of meteor/meteor#8414 and is normally
not necessary as part of the `meteor publish-release` process.
In order to allow for a blank `app.js` which occurs in the case of
a Meteor app using a fully-"package"-based structure with no actual
application code in the top-level. See meteor/meteor#8414 for more.
The error messages which come from UglifyJS tend to be quite cryptic, as
seen in issues like meteor/meteor#8370 or meteor/meteor#8020. The file,
line, and column are provided, however the message is garbled and the
stacktrace long and acutely harrowing. Since these errors are occurring
on automatically concatenated files, even the line number is sometimes
not helpful. Additionally, sourceMaps are not available in production
builds, intentionally. (I wasn't able to access them from
`file.getSourceMap()` or `file.sourceMap` at all.)
In addition to actually providing the name of the encapsulating file,
which provides _some_visibility, this commit implements a parser around
the UglifyJS error which detects the error and location information of
the error, seeks to the line in the concatenated source, reads the
inline filename, and provides it in the output.
Crude, but incredibly helpful in diagnosing this problem until a better
solution is reached.
