summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--entries/cffi-alpm-a-new-hope.rst3
-rw-r--r--metadata.json17
2 files changed, 18 insertions, 2 deletions
diff --git a/entries/cffi-alpm-a-new-hope.rst b/entries/cffi-alpm-a-new-hope.rst
new file mode 100644
index 0000000..3db414b
--- /dev/null
+++ b/entries/cffi-alpm-a-new-hope.rst
@@ -0,0 +1,3 @@
+Pyalpm has some annoyances to it, like leaking a lot of memory, and being a CPython module. It has many times been suggested that it should be re-written using CFFI or something similar though, and tonight I’ve been pretending that I don’t have anything better to do, so I started playing with CFFI and libalpm a bit.
+
+The result of my experimentations is available at `cffi-alpm <https://github.com/kyrias/cffi-alpm>`_, though it’s not particularly useful yet. Right now it has a class that wraps `alpm_list_t` and implements the Python `iterator protocol <https://www.python.org/dev/peps/pep-0234/>`_ for it, and have written a small example that searches the local DB. It’s not really much at all, but I do feel like I have a better feel of both CFFI and libalpm now, and might actually end up writing a complete pyalpm replacement.
diff --git a/metadata.json b/metadata.json
index 08426f5..422bf63 100644
--- a/metadata.json
+++ b/metadata.json
@@ -9,12 +9,25 @@
"title": "New blog!",
"date": "2016-11-12",
"file": "new-blog",
- "tags": ["blog"]
+ "tags": [
+ "blog"
+ ]
},
{
"title": "Style",
"date": "2016-11-15",
"file": "style",
- "tags": ["blog"]
+ "tags": [
+ "blog"
+ ]
+ },
+ {
+ "title": "cffi-alpm: A New Hope",
+ "date": "2016-11-17",
+ "file": "cffi-alpm-a-new-hope",
+ "tags": [
+ "python",
+ "libalpm"
+ ]
}
]