From:  Paul Irish <>
Date:  17 Jul 2013 08:06:48 Hong Kong Time

Re: Allow utilizing fragments of the minified file name for map file names


This was raised back on

IMO, the use case that drove this idea was human error between jQuery and
the Google CDN team. It's since been resolved.

I think a datauri sourcemap is a strong solution here to name changes.
Baking sugar into the spec that translates into many MB/s of overhead (in
aggregate) doesn't seem worth it when build processes could be followed

On Fri, Jul 12, 2013 at 6:51 AM, John Lenz  wrote:

> I would like to hear what the Chrome Dev folks think about this.  I think
> it is reasonable that the //# sourceUrl be required for eval code to work
> with this.
> On Wed, Jul 10, 2013 at 2:20 PM, Fitzgerald, Nick  > wrote:
>> On 7/9/13 9:25 AM, Michał Gołębiowski wrote:
>>> Ping again? We keep getting hurt because of that in jQuery, e.g. this
>>> CDN has a wrong location of the minified file:
>>> Could we have the `@hasImplicitSourceMapUrl` added to the spec, please?
>> I'm ok with adding `//# hasImplicitSourceMapURL` although (at the risk of
>> bike shedding) I think `//# implicitSourceMap` is a better name.
>> To find the source map for the script you just append ".map" to the
>> script's URL?
>> For inline scripts in a page do you still just append the ".map"? How
>> would this work with
>> doesn't work, would we do
>> Disallow implicit source map linking for inline scripts? What about eval'd
>> code?
>> Can you (or someone else) write up a proposal that answers these
>> questions and gives a solid algorithm for finding an implicit source map
>> that we would be able to just drop into the spec?
>> Nick
>> ______________________________**_________________
>> dev-js-sourcemap mailing list
>> dev-js-sourcemap@lists.**