summaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)AuthorFilesLines
2014-01-12Developer documentationTodd Gamblin6-5/+292
2014-01-12Remove trailing whitespace from license textTodd Gamblin93-469/+465
2014-01-11Site configuration docsTodd Gamblin2-0/+122
2014-01-09Add line to disable jekyll on github.Todd Gamblin1-0/+5
2014-01-09Fix docs to account for license header.Todd Gamblin2-0/+2
2014-01-09Add a rule for automatically creating gh-pages from html docs.Todd Gamblin1-0/+14
2014-01-09make empty _static directory "exist" to gitTodd Gamblin1-0/+4
Sphinx really wants the _static directory to exist, but git doesn't. This is a compromise to avoid superfluous warnings from sphinx. See here: http://stackoverflow.com/questions/115983/how-do-i-add-an-empty-directory-to-a-git-repository
2014-01-08Add release number to docs and license.Todd Gamblin1-1/+1
2014-01-08Change to github URL in docs.Todd Gamblin2-7/+9
2014-01-08Add license to python files and spack script.Todd Gamblin93-0/+2232
2014-01-07Change multimethods to take first match instead of being rigid.Todd Gamblin5-46/+71
- Formerly required that one and only one spec match - Now allows first match in a list (more flexible and more intuitive) - introduces new bug that provides() doesn't do the correct thing when a version is not in a range that has been explicitly provided. - TODO: fix this.
2014-01-06Packaging docs are mostly done.Todd Gamblin1-38/+93
2014-01-06More packaging docsTodd Gamblin2-73/+277
2014-01-06Local specs now called "anonymous specs"Todd Gamblin6-92/+126
- anonymous specs have no name - "local spec" came from these first being used in multimethods, i.e. "name of the local Package" - not the most intuitive name.
2014-01-06Simpler prefix args.Todd Gamblin8-10/+10
2014-01-06Remove install methods from mock packages (not needed)Todd Gamblin5-44/+5
2014-01-06SPACK-14: Bugfix in Spec.normalize()Todd Gamblin6-32/+107
- Normalize now updates the provider index as it addes package dependencies. - Fixes problem where this breaks: a depends_on mpi a depends_on b b depends_on mpich - Packages now restrict the mpi dependency to mpich
2014-01-04Change deps on 'mpich' to 'mpi'Todd Gamblin2-2/+2
2014-01-04More packaging documentationTodd Gamblin2-13/+477
2014-01-03Bug fix in how class names are determinedTodd Gamblin1-2/+5
2013-12-26More packaging documentation.Todd Gamblin10-62/+509
2013-12-25install() now takes spec AND prefixTodd Gamblin21-62/+122
2013-12-25Some fixes recommended by PyCharmTodd Gamblin13-25/+27
2013-12-25Starting on packaging docs.Todd Gamblin3-77/+98
2013-12-24make create and checksum consistent.Todd Gamblin6-97/+183
- create now searches and prompts for checksums. - makes package creation easier
2013-12-23spack checksum works.Todd Gamblin3-25/+56
2013-12-23spack edit fixed.Todd Gamblin2-7/+1
2013-12-23Spack create works again w/new package format.Todd Gamblin5-10/+75
2013-12-23Another multimethod test for vdepsTodd Gamblin2-9/+8
2013-12-23constrain() now consistent with satisfies()Todd Gamblin5-32/+103
- Added checks to constrain() so that it is consistent with satisfies() - Added many more test cases for satisfiability and constraints on deps - Virtual packages are handled properly in satisfies() and constrain() - bugfix: mpileaks^mpich2 would satisfy mpileaks^mpi@3: - this case is now handled.
2013-12-22Fix for SPACK-13, and satisfies() now handles deps.Todd Gamblin11-121/+419
Added more test cases for multimethods. In doing so, (re)discovered that satisfies() really needs to handle dependencies properly. Implemented support for dependencies in satisfies, but constrain() now isn't consistent (as we do not currently constrain deps), so need to implement that. Virtual dependency support probably needs some deeper thought. i.e., there is probably an intermediate DAG form that would make the needed checks easier. Right now we have to build ProviderIndexes to figure out how virtual dependencies are set up. If the vdep were preserved in the DAG, then we could just check for things like incompatible providers directly.
2013-12-21SPACK-2: Multimethods for specs.Todd Gamblin11-220/+363
- multi_function.py -> multimethod.py - Added @when decorator, which takes a spec and implements matching for method dispatch - Added multimethod unit test, covers basic cases.
2013-12-21More docs.Todd Gamblin3-8/+15
2013-12-21Color tweaks for find.Todd Gamblin2-22/+47
2013-12-21Docs for spack findTodd Gamblin1-7/+90
2013-12-21Made formatting specs easier with format() syntax.Todd Gamblin2-32/+77
- Replaced existing str_no_deps() function with more flexible format() function. - Spec.tree() can take a format argument now, as well.
2013-12-21SPACK-1: Multi-version installation now works front to back with specs.Todd Gamblin40-235/+617
2013-12-18Make dirty installs part of package lifecycleTodd Gamblin1-1/+1
2013-12-18More on installing with virtual packagesTodd Gamblin1-0/+29
2013-12-18Added a target to upload docs automatically.Todd Gamblin1-0/+3
2013-12-18Finished basic usage.Todd Gamblin15-167/+512
2013-12-12Start of basic documentationTodd Gamblin22-129/+944
2013-12-12Better info command.Todd Gamblin2-0/+38
2013-12-09Documentation skeletonTodd Gamblin19-0/+1340
2013-12-09use get instead of setdefault for kwargs.Todd Gamblin1-4/+4
2013-12-09type(foo) == bar -> isinstance(foo, bar)Todd Gamblin5-19/+21
Changed type checks in most places.
2013-12-08Additional circular reference checkingTodd Gamblin1-4/+8
2013-12-08Fixes for virtual packages.Todd Gamblin14-117/+393
- Added more tests - cleaned up spec preorder traversal - fixed concretization
2013-12-07Better test output -- include totals.Todd Gamblin2-24/+55
2013-12-07Added support for virtual dependencies ("provides")Todd Gamblin10-155/+533