"But now that you've run Maven again, what version of
svg.js got cached? The svg.js in my cache does need patching."
I think we crossed email replies because I already added comments about
that -see my previous post if you did not already.
To verify, I copied the downloaded and patched version of svg.js into the
cache, overwriting the last cached version, which was unpatched, and I can
repro the exact same error that I had before. I can only assure you I *did
not* do this originally :)
What I suggested is speculation, but I cannot see how else this could
happen.
So I think the resolution should either be to a) avoid using the download
plugin file cache somehow - or b) to patch a copy of the downloaded files
in a different temp location other than the original download location. I'm
afraid I don't know how to do either in maven at this point.
For now I am happy to find something that seems to 'fix' the issues I was
having with maven typedefs build. fwiw I also had a similar issue with the
googlemaps patch (only during recently builds) , it was the same type of
thing and was no longer apparent after emptying the same file cache.
On Thu, Jun 15, 2017 at 12:07 PM, Alex Harui <aha...@adobe.com.invalid>
wrote:
> That's interesting. But now that you've run Maven again, what version of
> svg.js got cached? The svg.js in my cache does need patching.
>
> -Alex
>
> On 6/14/17, 4:54 PM, "Greg Dove" <greg.d...@gmail.com> wrote:
>
> >I just manually emptied the file cache in .m2\repository\.
> >cache\maven-download-plugin
> >
> >And typedefs built fine via maven.
> >So my guess is that the download plugin is caching the files after they
> >have been patched and next time it is 'downloading' the patched file
> >
> >
> >On Thu, Jun 15, 2017 at 11:40 AM, Greg Dove <greg.d...@gmail.com> wrote:
> >
> >> A possible clue:
> >>
> >> I can see the patched file in :
> >> C:\Users\Greg\.m2\repository\.cache\maven-download-plugin
> >>
> >>
> >>
> >> On Thu, Jun 15, 2017 at 11:25 AM, Greg Dove <greg.d...@gmail.com>
> wrote:
> >>
> >>> Alex, fyi if I delete the target directory inside flex-typedefs (to be
> >>> sure it is 'clean', although I am quite sure mvn clean does this also)
> >>>,
> >>> then run mvn clean compile, i get the following in svg.js inside
> >>>downloads:
> >>>
> >>> /**
> >>> * @param {string} type
> >>> * @param {!EventListener|(function(!Event): (boolean|undefined))|
> >>>null}
> >>> listener
> >>> * @param {boolean=} opt_useCapture
> >>> */
> >>> SVGElementInstance.prototype.addEventListener = function(type,
> >>>listener,
> >>> opt_useCapture){};
> >>>
> >>> so for me at least, it does already seem to have the
> >>>"(function(!Event):"
> >>> etc in it.
> >>>
> >>>
> >>>
> >>>
> >>> On Thu, Jun 15, 2017 at 11:09 AM, Alex Harui <aha...@adobe.com.invalid
> >
> >>> wrote:
> >>>
> >>>> I just ran the curl command (sorry if my Outlook has munged the URL).
> >>>> I
> >>>> got the same hash. When I open it up, I see:
> >>>>
> >>>> /**
> >>>> * @param {string} type
> >>>> * @param {!EventListener|(function(Event): (boolean|undefined))|
> >>>>null}
> >>>> listener
> >>>> * @param {boolean=} opt_useCapture
> >>>> */
> >>>> SVGElementInstance.prototype.addEventListener = function(type,
> >>>>listener,
> >>>> opt_useCapture){};
> >>>>
> >>>> The patch is trying to replace
> >>>>
> >>>> (function(Event):
> >>>>
> >>>>
> >>>> With:
> >>>>
> >>>> (function(!Event):
> >>>>
> >>>>
> >>>> I verified locally that both my Ant and Maven builds from the repos
> >>>>can
> >>>> successfully download and patch svg.js. I did a clean before each
> >>>>build
> >>>> to verify that svg.js was not in the repo working copy folders.
> >>>>
> >>>>
> >>>> So you are saying that when you use curl to grab that file it already
> >>>>has
> >>>> "(function(!Event):" in it? That isn't the case for me. That is
> >>>>really
> >>>> strange given the hash was the same.
> >>>>
> >>>> Very puzzled,
> >>>> -Alex
> >>>>
> >>>>
> >>>> On 6/13/17, 7:23 PM, "Justin Mclean" <jus...@classsoftware.com>
> wrote:
> >>>>
> >>>> >Hi,
> >>>> >
> >>>> >Changing to -1 (binding).
> >>>> >
> >>>> >I can confirm the tyoedef issue is not due to differences in git
> >>>> version.
> >>>> >
> >>>> >My guess (but not confirmed) is that some people may to have the js
> >>>>file
> >>>> >file cached and that it has changed recently so the patch no longer
> >>>> >applies.
> >>>> >
> >>>> >I can confirm that the current js.patch will not apply to the file
> >>>> >current file here [1].
> >>>> >
> >>>> >Can some mind doing a diff with their version of the file and the
> >>>> version
> >>>> >here [1]?
> >>>> >
> >>>> >Or posting a hash of the file they have:
> >>>> >$curl
> >>>> >https://na01.safelinks.protection.outlook.com/?url=https%3A
> >>>> %2F%2Fstorage.g
> >>>> >oogleapis.com%2Fgoogle-code-archive-downloads%2Fv2%2Fcode.google.com
> >>>> %2Fclo
> >>>> >sureidl%2Fsvg.js&data=02%7C01%7C%7C2663b54b26d74c4e011c08d4
> >>>> b2cc5b6c%7Cfa7b
> >>>> >1b5a7b34438794aed2c178decee1%7C0%7C0%7C636330038163199802&s
> >>>> data=GzJiK3o9QT
> >>>> >3TcUiBFDUpKJIYELPlBIWbKHVsuIeIzxs%3D&reserved=0 > svg.js
> >>>> > % Total % Received % Xferd Average Speed Time Time
> >>>>Time
> >>>> >Current
> >>>> > Dload Upload Total Spent
> >>>>Left
> >>>> >Speed
> >>>> >100 155k 100 155k 0 0 74173 0 0:00:02 0:00:02
> >>>>--:--:--
> >>>> >74189
> >>>> >$md5 svg.js
> >>>> >MD5 (svg.js) = 568eb2c11ba33da58eb0471beef76cb9
> >>>> >
> >>>> >Thanks,
> >>>> >Justin
> >>>> >
> >>>> >1.
> >>>> >https://na01.safelinks.protection.outlook.com/?url=https%3A
> >>>> %2F%2Fstorage.g
> >>>> >oogleapis.com%2Fgoogle-code-archive-downloads%2Fv2%2Fcode.google.com
> >>>> %2Fclo
> >>>> >sureidl%2Fsvg.js&data=02%7C01%7C%7C2663b54b26d74c4e011c08d4
> >>>> b2cc5b6c%7Cfa7b
> >>>> >1b5a7b34438794aed2c178decee1%7C0%7C0%7C636330038163199802&s
> >>>> data=GzJiK3o9QT
> >>>> >3TcUiBFDUpKJIYELPlBIWbKHVsuIeIzxs%3D&reserved=0
> >>>> >
> >>>>
> >>>>
> >>>
> >>
>
>
