Merge branch 'master' of https://github.com/scala/scala-lang

heathermiller · heathermiller · commit 255d4a6f36b8 · 2016-04-05T17:01:35.000+02:00
diff --git a/blog/_posts/2016-03-14-announcing-the-scala-center.md b/blog/_posts/2016-03-14-announcing-the-scala-center.md
@@ -12,7 +12,7 @@ improve the language, its tools and its documentation. It will also help develop
 learn the language through massive open online courses, seminars and other activities.
 
 
-## Why A Center?
+## Why a Center?
 
 Scala has seen a steady rise in adoption. This is reflected in the
 [number of open Scala jobs](http://www.indeed.com/jobtrends/scala.html) and in Google's
@@ -21,15 +21,15 @@ In the [Redmonk rankings](http://sogrady-media.redmonk.com/sogrady/files/2016/02
 the language currently ranks 14th. Overall, we estimate that more than
 half a million developers are using Scala.
 
-Scala is the technology of choice for vibrant open-source ecosystem,
+Scala is the technology of choice for a vibrant open-source ecosystem,
 including popular platforms and frameworks such as
 [Spark](http://spark.apache.org/), [Kafka](http://kafka.apache.org/),
 [Akka](akka.io), [Finagle](https://twitter.github.io/finagle),
 [Cats](http://typelevel.org/cats/), and
 [Play](https://www.playframework.com/), to name just a few.
 
 The large user base and the diversity of projects built on Scala mean
-that there are many demands on the language and its tool set. There is also a large pool of capable and motivated contributors. The first mission of the Scala center is to consult with
+that there are many demands on the language and its tool set. There is also a large pool of capable and motivated contributors. The first mission of the Scala Center is to consult with
 the user community, define common goals, and organize contributions
 and its own resources to achieve these goals.
 
@@ -39,7 +39,7 @@ from other languages and requires some conscious investment in
 learning the language to reap most of it benefits. Learning Scala, if
 done right, is much more than just internalizing syntax. It usually
 deeply affects the way developers think about their activity. We have
-pioneered open online education, with more than 400'000 registrations
+pioneered open online education, with more than 400,000 registrations
 to our MOOCs on Coursera. The Scala Center will take over and enlarge
 these efforts, integrating them in a specialization that teaches
 functional programming and data analysis using Scala and Spark.
@@ -50,17 +50,17 @@ The Scala Center is a designated unit at EPFL, independent from the
 Scala research group (LAMP). Its first two officers are executive director Heather Miller,
 and principal engineer Jon Pretty. The Center is funded by EPFL and a group of industrial
 members. The initial industrial founding members of the center are:
-Goldman Sachs, IBM, Lightbend, Nitro, Verizon. We are keen to
+Goldman Sachs, IBM, Lightbend, Nitro, and Verizon. We are keen to
 welcome other members in the future.
 
-## Relationship with the Community
+## Relationship with the community
 
 The Center aims in helping and complementing existing entities in the
 Scala ecosystem. In particular:
 
 Lightbend (formerly Typesafe) will continue to maintain and evolve the
 stable Scala distribution. Lightbend is a founding member of the
-center and its engineers are looking forward to participate in its
+center and its engineers are looking forward to participating in its
 projects.
 
 Existing open source projects and platforms will work as before. The
@@ -70,14 +70,16 @@ of projects of common interest, subject to its available resources.
 
 One specific project where it looks like we have enough interest and
 pledges of funding is [Scala.js](http://www.scala-js.org/).  Scala.js
-is developed so far by my lab at EPFL (specifically Sebastien Doeraene
+is developed so far by the lab at EPFL (specifically Sébastien Doeraene
 and Nicolas Stucki) with the help of outside contributors, including Lightbend.
 The Scala Center should be a
 logical platform to put this great work on a more stable basis.
 
-## Please Contribute!
+## Please contribute!
 
-We are starting out now with a rough direction and expect that most details of the Center’s working will become clearer once we have gained some experience in our activities. For now, we are soliciting the community for help. If you have time or funding to contribute, and would like to work on a project that is useful to the community at large, please join us in the [Scala Center gitter channel](https://gitter.im/scala/center) and get involved!
+We are starting out now with a rough direction and expect that most details of the Center’s working will become clearer once we have gained some experience in our activities. For now, we are soliciting the community for help.
+
+If you have time or funding to contribute, and would like to work on a project that is useful to the community at large, please join us in the [Scala Center gitter channel](https://gitter.im/scala/center) and get involved!
 
 
 
diff --git a/blog/_posts/2016-03-17-getting-into-scaladoc-dev.md b/blog/_posts/2016-03-17-getting-into-scaladoc-dev.md
@@ -0,0 +1,127 @@
+---
+layout: blog
+post-type: blog
+by: Felix Mulder
+title: Getting into Scaladoc Development
+disqus: true
+---
+
+Over the past months I've been working on the new Scaladoc - and it's coming
+along nicely. Together with
+[@heathercmiller](http://twitter.com/heathercmiller) and
+[@vladureche](http://twitter.com/vladureche) we've overhauled the overall look
+of Scaladoc as well as having added a bunch of useful features - like complete
+member search!
+
+With the new Scala Center recently announced, one of the things discussed has
+been how to lower the entry barrier on important open source projects. This blog
+post aims to do exactly that for Scaladoc. There are a lot of things we want
+to do with the docs - and we'd love help! If that piques your interest - read on.
+
+## Scaladoc for 2.12.x ##
+The new major version of Scala is scheduled for later this year, and currently
+its doc tool generates sites like this one:
+
+![Scala Collections](http://i.imgur.com/TOkD4JF.png)
+
+The new search functionality allows you to not only look for so called
+"top-level entities" (`class`, `trait`, `object` and `package`) but members like
+methods and even `val` and `var`s. The new search is not at all taxing, on a
+big library like the collections - the results are shown almost
+instantaneously. Nevertheless, we've got a progress bar showing you something's
+actually happening:
+
+![Scala Collections](http://imgur.com/jLxcRpx.png)
+
+## How does Scaladoc work? ##
+The Scaladoc tool is quite easy to understand - there are basically three things
+that happen once you've specified what you want documented:
+
+1. Scaladoc compiles the sources using the first step of the compiler (the
+   frontend). This will fill a tree structure known as `Universe` with all the
+   information about the sources (classes, members, doc string etc).
+
+   **See:** [DocFactory.scala](https://github.com/scala/scala/blob/2.12.x/src/scaladoc/scala/tools/nsc/doc/DocFactory.scala)
+
+2. Copy all needed assets to out location (i.e. scripts and CSS-files)
+
+   **See:** [HtmlFactory.scala](https://github.com/scala/scala/blob/2.12.x/src/scaladoc/scala/tools/nsc/doc/html/HtmlFactory.scala).
+
+3. Traverse the `Universe` tree and for each top-level entity create an
+   HTML-page using the `Entity` class.
+
+   **See:** [Entity.scala](https://github.com/scala/scala/blob/2.12.x/src/scaladoc/scala/tools/nsc/doc/html/page/Entity.scala)
+
+The bulk of these steps are carried out inside the `HtmlFactory` class in
+[HtmlFactory.scala](https://github.com/scala/scala/blob/2.12.x/src/scaladoc/scala/tools/nsc/doc/html/HtmlFactory.scala).
+Have a look there for a more complete overview.
+
+## How do I generate the docs? ##
+
+Simple:
+
+    $ git clone git@github.com:scala/scala.git && cd scala
+    $ ant docs # "ant docs.lib" if you just want the standard library
+
+## Where to begin ##
+
+The markup of the entity page is defined in
+[Entity.scala](https://github.com/scala/scala/blob/2.12.x/src/scaladoc/scala/tools/nsc/doc/html/page/Entity.scala).
+If you're looking to change the HTML that's where you will want to look first.
+
+If you want to add some static asset, remember that `HtmlFactory` needs to copy
+it to the destination. So be sure to add your new resource to `libResources` in
+[HtmlFactory.scala](https://github.com/scala/scala/blob/2.12.x/src/scaladoc/scala/tools/nsc/doc/html/HtmlFactory.scala).
+
+All the current static assets are located in the
+[lib](https://github.com/scala/scala/tree/2.12.x/src/scaladoc/scala/tools/nsc/doc/html/resource/lib) directory.
+That's where you'd want to put new things (it is also where `index.js` etc live!)
+
+**How can I see my changes?** You have two options, if you changed the markup -
+you'll need to regenerate the docs:
+
+    $ ant docs.clean && ant docs
+
+When the changes are to scripts or style sheets I use a small `Makefile` in the
+root of the project. This makefile will simply copy the resources to the same location
+as the generated docs:
+
+    all:
+        cp src/scaladoc/scala/tools/nsc/doc/html/resource/lib/{index.css,index.js,template.css,template.js,diagrams.css,diagrams.js} build/scaladoc/library/lib/
+
+Regenerating the docs all the time is a pain - if you just want to generate
+them for a single file, do this:
+
+    $ ant quick.bin
+    $ build/quick/bin/scaladoc path/to/source/File.scala
+
+Lastly, before you open your pull request - test your changes!
+
+    $ ant test.scaladoc
+
+## Where to really begin ##
+
+There's still a lot of things to do before this can be considered release ready.
+Here's a laundry list of things we need to do:
+
+* General
+    - Async all the things! We should be able to parallelize the generation of
+      the doc pages
+    - CSS cleanup! There are a lot of things not necessary anymore, these could
+      simply be deleted
+* Mobile, the docs look decent on mobile - but there's still a ton to do!
+    - Landscape kills some of our layout
+    - There is no good package overview (list on the right) - ideas welcome!
+* Desktop
+    - Keyboard navigation on entity pages (i.e. between members)
+    - Better keyboard navigation on search results
+    - Minor layout issues
+    - Showing existence of companion `class`/`trait`/`object` (maybe *"You
+      won't believe who* `Companion object Boolean` *is going home with
+      tonight!"*)
+
+The full list of ideas and things to do can be found on
+[scala-dev](https://github.com/scala/scala-dev/issues/84). Comment on the issue
+or reach out to one of us (@felixmulder, @heathermiller, @vladureche) on the
+[scala/contributors gitter](https://gitter.im/scala/contributors) if you're
+interested in knowing more!
diff --git a/gsoc/2016.md b/gsoc/2016.md
@@ -208,6 +208,56 @@ construction.** Ideally, it should point to a compiler project you have written
 
 Supervised by [@sjrd](https://github.com/sjrd/)
 
+### Scala Compiler Plugin for Parser Combinators
+
+Parsing can be found everywhere in computing today: from log analysis to
+computer languages. Functional programming offers concise approach to
+parser generators development: a parsing grammar for generator is composed
+via higher-order functions -- [parser combinators](http://www.cs.nott.ac.uk/~pszgmh/monparsing.pdf).
+
+As a program, parser generator should be programmed easily and work fast.
+Popular parser generators in Scala ([Scala Combinator Parsers](https://github.com/scala/scala-parser-combinators),
+[parboiled2](http://parboiled2.org), [FastParse](https://github.com/lihaoyi/fastparse ),
+[PapaCarlo](http://lakhin.com/projects/papa-carlo/), etc.) can be much faster
+or have complications in development.
+
+Scala Combinator Parsers library is most concise. It is implemented as plain
+library in Scala. And because of that has few facilities for optimizations.
+
+parboiled2 drops off most of algebraic sanity, and leaves a programmer with
+imperative-style parsing and Value Stack. That causes type soundness bugs,
+clumsy code that is hard to debug, and bad error reporting. FastParse fixes
+most critical issues of parboiled2. But both lacks can’t perform cross-rules
+analysis and deep optimisations.
+
+The proposed solution requires users to use SBT build system with external plugin.
+Scala compiler plugin should rewrite Scala AST of a parser: compile high-level
+concise monadic Scala Combinator Parsers to highly-effective runtime code.
+
+The plugin would be written in Scala. Existing projects written in Scala
+are expected. Ideally you should have understanding of compilers in
+general and Scala compiler inner parts.
+
+Supervised by [@alexander-myltsev](https://github.com/alexander-myltsev),
+[a core contributor to parboiled2 during GSoC'2013](https://github.com/sirthias/parboiled2/commits?author=alexander-myltsev)
+and [biological names parser based or parboiled2](https://github.com/GlobalNamesArchitecture/gnparser/commits?author=alexander-myltsev)
+writing [notes sometimes in his blog](http://myltsev.com/tags/#parboiled2).
+
+### New XML for Scala
+
+This project is going to be about new platform-independent (i.e. cross-compile to both JVM
+and Scala.js) implementation of XML library based on fastparse parser combinators.
+
+- Finish the XML 1.0 parser
+- Refine the AST model
+- Implement quasiquotes based on the parser and AST model
+
+Original prototype is available at:
+
+[https://github.com/densh/scala-nxml](https://github.com/densh/scala-nxml)
+
+Supervised by [@densh](https://github.com/densh)
+
 ## ENSIME project ideas
 
 ENSIME is a JVM process that indexes your dependencies and understands your source code
