2.x: Correct marble diagram dimensions

[Stephan202]

This fixes part of #4535, as requested in #5067.

These changes were generated using the following script:

#!/usr/bin/env bash

find . -name '*.java' -print0 \
  | xargs -0 grep '^\s*\*' \
  | grep -oP 'https.*?\.png' \
  | sort -u \
  | while read -r I; do
      echo "Updating ${I}..."
      # Gather image details.
      DATA="$(identify <(curl -f -L -s "${I}"))"

      # Extract the image's true width and height.
      WIDTH="$(echo "${DATA}" | grep -oP '(?<=PNG )\d+')"
      HEIGHT="$(echo "${DATA}" | grep -oP 'PNG \d+x\d+' | grep -oP '(?<=x)\d+')"

      # Determine the image's scaled height, assuming a width of 640 pixels.
      NEW_HEIGHT="$(echo 'scale=100; s='"${WIDTH}"' / 640; scale=0; '"${HEIGHT}"' / s' | bc)"

      # Update all `<img/>` tags referencing the image.
      find . -name '*.java' -print0 \
        | xargs -0 perl -i -p -e 's,<img( (width|height)="[^"]*")* src="\Q'"${I}"'\E",<img width="640" height="'"${NEW_HEIGHT}"'" src="'"${I}"'",g'
    done

[Stephan202]

(Note that I assumed that it's okay to just fix the width of all images at 640px; that seemed to be the prevailing style.)

[akarnokd]

Great scripting job!

[Stephan202]

Thanks! :). Since the change only touches JavaDoc, I assume the Travis CI job failure is spurious/unrelated? Otherwise I'll need a pointer on how to correct the situation.

[akarnokd]

The failure is due to time sensitive tests failing on an overloaded CI. I'll try to resolve it separately, don't worry about it.

[codecov]

Codecov Report

Merging #5087 into 2.x will increase coverage by 0.07%.

@@             Coverage Diff              @@
##                2.x    #5087      +/-   ##
============================================
+ Coverage     95.51%   95.59%   +0.07%     
- Complexity     5547     5550       +3     
============================================
  Files           614      614              
  Lines         39580    39580              
  Branches       5559     5559              
============================================
+ Hits          37805    37836      +31     
+ Misses          778      758      -20     
+ Partials        997      986      -11

Impacted Files	Coverage Δ	Complexity Δ
.../java/io/reactivex/processors/ReplayProcessor.java	98.75% <ø> (+0.2%)	49 <ø> (ø)	❌
...ain/java/io/reactivex/subjects/PublishSubject.java	97.87% <ø> (ø)	40 <ø> (ø)	❌
...ernal/operators/flowable/BlockingFlowableNext.java	95.16% <ø> (ø)	2 <ø> (ø)	❌
src/main/java/io/reactivex/Flowable.java	100% <ø> (ø)	520 <ø> (ø)	❌
...main/java/io/reactivex/subjects/ReplaySubject.java	97.11% <ø> (ø)	51 <ø> (ø)	❌
...ators/observable/BlockingObservableMostRecent.java	100% <ø> (ø)	2 <ø> (ø)	❌
...in/java/io/reactivex/subjects/BehaviorSubject.java	89.63% <ø> (+4.66%)	55 <ø> (-1)	❌
...l/operators/observable/BlockingObservableNext.java	98.36% <ø> (ø)	2 <ø> (ø)	❌
...n/java/io/reactivex/parallel/ParallelFlowable.java	100% <ø> (ø)	41 <ø> (ø)	❌
src/main/java/io/reactivex/Completable.java	100% <ø> (ø)	112 <ø> (ø)	❌
... and 43 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8819cc9...066f8e3. Read the comment docs.