ipums · renae-r · Apr 13, 2026 · Feb 2, 2026 · Feb 4, 2026 · Feb 4, 2026
diff --git a/.github/workflows/coverage.yml b/.github/workflows/coverage.yml
@@ -17,7 +17,7 @@ jobs:
       - name: Setup Python
         uses: actions/setup-python@v5
         with:
-          python-version: 3.9
+          python-version: 3.11
 
       - name: Install poetry
         uses: snok/install-poetry@v1

diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
@@ -13,7 +13,7 @@ jobs:
     strategy:
       fail-fast: true
       matrix:
-        python-version: ["3.9", "3.10", "3.11", "3.12"]
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
 
     steps:
       - uses: actions/checkout@v4

diff --git a/docs/source/change-log.rst b/docs/source/change-log.rst
@@ -11,6 +11,18 @@ This project adheres to `Semantic Versioning`_.
 .. _Semantic Versioning: http://semver.org/
 
 
+0.8.0
+-----
+2026-02-?
+
+* New Features
+
+  * Support for IPUMS DHS extract API
+
+* Breaking Changes
+
+  * Dropped support for Python 3.9
+
 0.7.0
 -----
 2025-06-09

diff --git a/docs/source/getting_started.rst b/docs/source/getting_started.rst
@@ -8,7 +8,7 @@ Getting Started
 Installation
 ------------
 
-This package requires that you have at least Python 3.9 installed.
+This package requires that you have at least Python 3.10 installed.
 
 Install with ``pip``:
 

diff --git a/docs/source/ipums_api/index.rst b/docs/source/ipums_api/index.rst
@@ -73,6 +73,11 @@ available features for all collections currently supported by the API:
       - ``mtus``
       - **X**
       - 
+    * - `IPUMS DHS <https://www.idhsdata.org/idhs/index.shtml>`__
+      - Microdata
+      - ``dhs``
+      - **X**
+      - 
     * - `IPUMS NHIS <https://nhis.ipums.org/nhis/>`__
       - Microdata
       - ``nhis``

diff --git a/docs/source/ipums_api/ipums_api_micro/index.rst b/docs/source/ipums_api/ipums_api_micro/index.rst
@@ -95,7 +95,7 @@ The table below shows the available data structures and the IPUMS data collectio
       - all IPUMS microdata collections
     * - hierarchical
       - ``data_structure={"hierarchical": {}}``
-      - all IPUMS microdata collections
+      - ``usa``, ``cps``, ``atus``, ``ahtus``, ``mtus``, ``nhis``, ``meps``, ``ipumsi``
     * - rectangular on Activity
       - ``data_structure={"rectangular": {"on": "A"}}``
       - ``atus``, ``ahtus``, ``mtus``
@@ -132,7 +132,7 @@ Extract Features
 ----------------
 
 Certain features of a :class:`MicrodataExtract<ipumspy.api.extract.MicrodataExtract>` can be 
-added or updated before an extract request is submitted. This section 
+added or updated before an extract request is submitted. Note that not all features are availale for every IPUMS data collection. For example, case selection is not a feature of IPUMS Time Use data and automatic inclusion of data quality flags is not a feature of IPUMS DHS. Details regarding available extract features for each IPUMS data collection can be found at the `IPUMS developer portal <https://developer.ipums.org/docs/v2/apiprogram/apis/microdata/>`_. This section 
 demonstrates adding features to the following IPUMS CPS extract.
 
 .. code:: python

diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "ipumspy"
-version = "0.7.0"
+version = "0.8.0"
 description = "A collection of tools for working with IPUMS data"
 authors = ["Kevin H. Wilson <kevin_wilson@brown.edu>",
            "Renae Rodgers <rodge103@umn.edu>"]
@@ -10,29 +10,29 @@ packages = [
 ]
 
 [tool.poetry.dependencies]
-python = "^3.9"
+python = "^3.10"
 pandas = "^2.2.3"
 numpy = "^2.0.0"
 click = "^8.0.0"
-pyarrow = "^18.1.0"
-requests = {extras = ["use_chardet_on_py3"], version = "^2.26.0"}
+pyarrow = "^23.0.0"
+requests = {extras = ["use_chardet_on_py3"], version = "^2.32.5"}
 importlib-metadata = "^4.13.0"
 PyYAML = "^6.0.1"
 
-[tool.poetry.dev-dependencies]
+[tool.poetry.group.dev.dependencies]
 black = "^24.4.0"
 pylint = "^2.7.4"
 isort = "^5.8.0"
 mypy = "^1.0.1"
 
 [tool.poetry.group.test.dependencies]
-pytest = "^8.3.4"
+pytest = "^9.0.0"
 pytest-cov = "^4.0.0"
 python-dotenv = "^1.0.0"
-fastapi = "^0.115.12"
-uvicorn = {extras = ["standard"], version = "^0.34.1"}
-pytest-recording = "^0.12.2"
-vcrpy = "^4.2.1"
+fastapi = "^0.128.1"
+uvicorn = {extras = ["standard"], version = "^0.40.0"}
+pytest-recording = "^0.13.4"
+vcrpy = "^8.0.0"
 
 [tool.poetry.group.docs]
 optional=true

diff --git a/src/ipumspy/api/extract.py b/src/ipumspy/api/extract.py
@@ -360,6 +360,7 @@ def _get_collection_type(collection: str) -> str:
         "mtus": "microdata",
         "nhis": "microdata",
         "meps": "microdata",
+        "dhs": "microdata",
         "nhgis": "aggregate_data",
         "ihgis": "aggregate_data",
     }
@@ -680,7 +681,7 @@ def build(self) -> Dict[str, Any]:
             }
 
         # XXX shoehorn fix until server-side bug is fixed
-        if self.collection == "meps":
+        if self.collection == "meps" or self.collection == "dhs":
             for variable in built["variables"].keys():
                 built["variables"][variable].pop("attachedCharacteristics")
 

diff --git a/src/ipumspy/readers.py b/src/ipumspy/readers.py
@@ -217,16 +217,11 @@ def _get_common_vars(ddi: ddi_definitions.Codebook, data_description: List):
     # these are delimited by spaces within the string attribute
     # this list would probably be a useful thing to have as a file-level attribute...
 
-    # XXX: this is to work around an issue with the Health Surveys DDI specifically.
-    # Revert to previous method of using the file_description rectypes once this
-    # DDI issue is fixed
-    rectype_desc = [desc for desc in data_description if desc.name == "RECTYPE"][0]
-    all_rectypes = rectype_desc.rectype.split(" ")
     common_vars = [
         desc.name
         for desc in data_description
-        # if sorted(desc.rectype.split(" ")) == sorted(ddi.file_description.rectypes)
-        if sorted(desc.rectype.split(" ")) == sorted(all_rectypes)
+        if sorted(desc.rectype.split(" ")) == sorted(ddi.file_description.rectypes)
+        # if sorted(desc.rectype.split(" ")) == sorted(all_rectypes)
     ]
 
     return common_vars
@@ -355,14 +350,8 @@ def read_hierarchical_microdata(
     else:
         df_dict = {}
         common_vars = _get_common_vars(ddi, data_description)
-        # XXX: this is to work around an issue with the Health Surveys DDI specifically.
-        # Revert to previous method of using the file_description rectypes once this
-        # DDI issue is fixed
-        rectype_desc = [desc for desc in data_description if desc.name == "RECTYPE"][0]
-        all_rectypes = rectype_desc.rectype.split(" ")
-
-        # for rectype in ddi.file_description.rectypes:
-        for rectype in all_rectypes:
+
+        for rectype in ddi.file_description.rectypes:
             rectype_vars = _get_rectype_vars(
                 ddi, rectype, common_vars, data_description
             )

diff --git a/tests/fixtures/meps_00020.dat.gz → tests/fixtures/meps_00045.dat.gz b/tests/fixtures/meps_00020.dat.gz → tests/fixtures/meps_00045.dat.gz