OLD | NEW |
(Empty) | |
| 1 # GYP vs. CMake # |
| 2 |
| 3 GYP was originally created to generate native IDE project files (Visual Studio,
Xcode) for building [Chromium](http://www.chromim.org). |
| 4 |
| 5 The functionality of GYP is very similar to the [CMake](http://www.cmake.org) |
| 6 build tool. Bradley Nelson wrote up the following description of why the team |
| 7 created GYP instead of using CMake. The text below is copied from |
| 8 http://www.mail-archive.com/webkit-dev@lists.webkit.org/msg11029.html |
| 9 |
| 10 ``` |
| 11 |
| 12 Re: [webkit-dev] CMake as a build system? |
| 13 Bradley Nelson |
| 14 Mon, 19 Apr 2010 22:38:30 -0700 |
| 15 |
| 16 Here's the innards of an email with a laundry list of stuff I came up with a |
| 17 while back on the gyp-developers list in response to Mike Craddick regarding |
| 18 what motivated gyp's development, since we were aware of cmake at the time |
| 19 (we'd even started a speculative port): |
| 20 |
| 21 |
| 22 I did an exploratory port of portions of Chromium to cmake (I think I got as |
| 23 far as net, base, sandbox, and part of webkit). |
| 24 There were a number of motivations, not all of which would apply to other |
| 25 projects. Also, some of the design of gyp was informed by experience at |
| 26 Google with large projects built wholly from source, leading to features |
| 27 absent from cmake, but not strictly required for Chromium. |
| 28 |
| 29 1. Ability to incrementally transition on Windows. It took us about 6 months |
| 30 to switch fully to gyp. Previous attempts to move to scons had taken a long |
| 31 time and failed, due to the requirement to transition while in flight. For a |
| 32 substantial period of time, we had a hybrid of checked in vcproj and |
| 33 gyp generated |
| 34 vcproj. To this day we still have a good number of GUIDs pinned in the |
| 35 gyp files, |
| 36 because different parts of our release pipeline have leftover assumptions |
| 37 regarding manipulating the raw sln/vcprojs. This transition occurred from |
| 38 the bottom up, largely because modules like base were easier to convert, and |
| 39 had a lower churn rate. During early stages of the transition, the majority |
| 40 of the team wasn't even aware they were using gyp, as it integrated into |
| 41 their existing workflow, and only affected modules that had been converted. |
| 42 |
| 43 2. Generation of a more 'normal' vcproj file. Gyp attempts, particularly on |
| 44 Windows, to generate vcprojs which resemble hand generated projects. It |
| 45 doesn't generate any Makefile type projects, but instead produces msvs |
| 46 Custom Build Steps and Custom Build Rules. This makes the resulting projects |
| 47 easier to understand from the IDE and avoids parts of the IDE that simply |
| 48 don't function correctly if you use Makefile projects. Our early hope with |
| 49 gyp was to support the least common denominator of features present in each |
| 50 of the platform specific project file formats, rather than falling back on |
| 51 generated Makefiles/shell scripts to emulate some common abstraction. CMake by |
| 52 comparison makes a good faith attempt to use native project features, but |
| 53 falls back on generated scripts in order to preserve the same semantics on |
| 54 each platforms. |
| 55 |
| 56 3. Abstraction on the level of project settings, rather than command line |
| 57 flags. In gyp's syntax you can add nearly any option present in a hand |
| 58 generated xcode/vcproj file. This allows you to use abstractions built into |
| 59 the IDEs rather than reverse engineering them possibly incorrectly for |
| 60 things like: manifest generation, precompiled headers, bundle generation. |
| 61 When somebody wants to use a particular menu option from msvs, I'm able to |
| 62 do a web search on the name of the setting from the IDE and provide them |
| 63 with a gyp stanza that does the equivalent. In many cases, not all project |
| 64 file constructs correspond to command line flags. |
| 65 |
| 66 4. Strong notion of module public/private interface. Gyp allows targets to |
| 67 publish a set of direct_dependent_settings, specifying things like |
| 68 include_dirs, defines, platforms specific settings, etc. This means that |
| 69 when module A depends on module B, it automatically acquires the right build |
| 70 settings without module A being filled with assumptions/knowledge of exactly |
| 71 how module B is built. Additionally, all of the transitive dependencies of |
| 72 module B are pulled in. This avoids their being a single top level view of |
| 73 the project, rather each gyp file expresses knowledge about its immediate |
| 74 neighbors. This keep local knowledge local. CMake effectively has a large |
| 75 shared global namespace. |
| 76 |
| 77 5. Cross platform generation. CMake is not able to generate all project |
| 78 files on all platforms. For example xcode projects cannot be generated from |
| 79 windows (cmake uses mac specific libraries to do project generation). This |
| 80 means that for instance generating a tarball containing pregenerated |
| 81 projects for all platforms is hard with Cmake (requires distribution to |
| 82 several machine types). |
| 83 |
| 84 6. Gyp has rudimentary cross compile support. Currently we've added enough |
| 85 functionality to gyp to support x86 -> arm cross compiles. Last I checked |
| 86 this functionality wasn't present in cmake. (This occurred later). |
| 87 |
| 88 |
| 89 That being said there are a number of drawbacks currently to gyp: |
| 90 |
| 91 1. Because platform specific settings are expressed at the project file |
| 92 level (rather than the command line level). Settings which might otherwise |
| 93 be shared in common between platforms (flags to gcc on mac/linux), end up |
| 94 being repeated twice. Though in fairness there is actually less sharing here |
| 95 than you'd think. include_dirs and defines actually represent 90% of what |
| 96 can be typically shared. |
| 97 |
| 98 2. CMake may be more mature, having been applied to a broader range of |
| 99 projects. There a number of 'tool modules' for cmake, which are shared in a |
| 100 common community. |
| 101 |
| 102 3. gyp currently makes some nasty assumptions about the availability of |
| 103 chromium's hermetic copy of cygwin on windows. This causes you to either |
| 104 have to special case a number of rules, or swallow this copy of cygwin as a |
| 105 build time dependency. |
| 106 |
| 107 4. CMake includes a fairly readable imperative language. Currently Gyp has a |
| 108 somewhat poorly specified declarative language (variable expansion happens |
| 109 in sometimes weird and counter-intuitive ways). In fairness though, gyp assumes |
| 110 that external python scripts can be used as an escape hatch. Also gyp avoids |
| 111 a lot of the things you'd need imperative code for, by having a nice target |
| 112 settings publication mechanism. |
| 113 |
| 114 5. (Feature/drawback depending on personal preference). Gyp's syntax is |
| 115 DEEPLY nested. It suffers from all of Lisp's advantages and drawbacks. |
| 116 |
| 117 -BradN |
| 118 ``` |
OLD | NEW |