Tweak read

.codecov.yml

@@ -1 +1,4 @@
-comment: false
+comment:
+  layout: "reach, diff, flags, files"
+  behavior: default
+  require_changes: true

.github/workflows/flake8.yml

@@ -0,0 +1,20 @@
+name: Flake8
+
+on:
+  pull_request:
+  push:
+    branches: [master]
+
+jobs:
+  flake8:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - name: Setup Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: '3.10'
+      - name: Install flake8
+        run: python -m pip install flake8 flake8-pyproject
+      - name: Run flake8
+        run: flake8

.github/workflows/main.yml

@@ -19,18 +19,49 @@ jobs:
       fail-fast: false
 
       matrix:
-        version: ['1.6', '1.7', '1.8', '1.9', '1.10']
+        version: ['1', 'lts']
         os: [ubuntu-latest, macos-latest]
 
     steps:
       - name: Checkout SegyIO
         uses: actions/checkout@v2
 
       - name: Setup julia
-        uses: julia-actions/setup-julia@v1
+        uses: julia-actions/setup-julia@v2
         with:
           version: ${{ matrix.version }}
           arch: x64
 
       - name: Run tests
-        uses: julia-actions/julia-runtest@master
+        uses: julia-actions/julia-runtest@latest
+        with:
+          annotate: true
+
+      - uses: julia-actions/julia-processcoverage@v1
+
+      - uses: codecov/codecov-action@v4
+        with:
+          file: lcov.info
+          token: ${{ secrets.CODECOV_TOKEN }}
+
+  python-tests:
+    name: Python tests
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - name: Setup Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: '3.10'
+      - name: Install package
+        run: python -m pip install -e .
+      - name: Install dependencies
+        run: python -m pip install pytest pytest-cov codecov
+      - name: Run tests with coverage
+        run: pytest -vs --cov=pysegy --cov-report=xml
+      - name: Upload coverage
+        uses: codecov/codecov-action@v4
+        with:
+          fail_ci_if_error: true
+          files: coverage.xml
+          token: ${{ secrets.CODECOV_TOKEN }}

.gitignore

@@ -1,3 +1,9 @@
 *.jl.cov
 *.jl.*.cov
 *.jl.mem
+__pycache__/
+pysegy/__pycache__/
+pysegy/tests/__pycache__/
+.coverage
+coverage.xml
+htmlcov/

Project.toml

@@ -1,6 +1,6 @@
 name = "SegyIO"
 uuid = "157a0f19-4d44-4de5-a0d0-07e2f0ac4dfa"
-authors = ["Henryk Modzelewski <henryk_modzelewski@mac.com>"]
+authors = ["Henryk Modzelewski <henryk_modzelewski@mac.com>", "Mathias Louboutin <mathias@devitocodes.com>"]
 version = "0.8.5"
 
 [deps]

README.md

@@ -17,3 +17,5 @@ SegyIO is a registered package and can be installed directly from the julia pack
 ## Extension
 
 SegyIO is implemented for POSIX systems. For Cloud storage, use [CloudSegyIO.jl](https://github.com/slimgroup/CloudSegyIO.jl), the Cloud storage extension of SegyIO.
+
+A minimal Python implementation of the core reader and writer is available in the `pysegy/` directory. Tests for this version run on GitHub Actions with coverage reporting via Codecov.

benchmarks/scan_benchmark.py

@@ -0,0 +1,20 @@
+import os
+import time
+import pysegy as seg
+
+DATA_DIR = os.path.join(os.path.dirname(__file__), '..', 'data')
+PATTERN = 'overthrust_2D_shot_*.segy'
+
+
+def main():
+    t0 = time.perf_counter()
+    seg.segy_scan(DATA_DIR, PATTERN, threads=1)
+    t1 = time.perf_counter()
+    seg.segy_scan(DATA_DIR, PATTERN)
+    t2 = time.perf_counter()
+    print(f"Sequential: {t1 - t0:.3f}s")
+    print(f"Threaded:   {t2 - t1:.3f}s")
+
+
+if __name__ == '__main__':
+    main()

pyproject.toml

@@ -0,0 +1,46 @@
+[build-system]
+requires = ["setuptools", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "pysegy"
+dynamic = ["version"]
+description = "Minimal Python implementation of SegyIO"
+authors = [
+  { name = "Mathias Louboutin",  email = "mathias@devitocodes.com" },
+]
+readme = "README.md"
+requires-python = ">=3.10"
+classifiers = [
+  "Programming Language :: Python :: 3",
+]
+license = { file = "LICENSE" }
+dependencies = [
+    "numpy>=2,<3",
+    "fsspec>=2024.3",
+    "cloudpickle>=3",
+]
+
+[project.optional-dependencies]
+dev = ["flake8", "flake8-pyproject"]
+
+[tool.setuptools_scm]
+fallback_version = "0+untagged"
+
+[tool.pytest.ini_options]
+addopts = "-v"
+testpaths = ["pysegy/tests"]
+
+[tool.flake8]
+max-line-length = 90
+ignore = [
+  "F403","E226","E731","E275",
+  "W503","F405","E722","E741",
+  "W504","W605"
+]
+exclude = [ ".github" ]
+
+[tool.setuptools.packages.find]
+where = ["."]
+include = ["pysegy*"]
+

pysegy/__init__.py

@@ -0,0 +1,52 @@
+"""Minimal Python port of SegyIO.jl."""
+# flake8: noqa
+
+import logging
+
+
+logger = logging.getLogger('pysegy')
+stream_handler = logging.StreamHandler()
+logger.addHandler(stream_handler)
+
+from .types import (
+    BinaryFileHeader,
+    BinaryTraceHeader,
+    FileHeader,
+    SeisBlock,
+)
+from .read import (
+    read_fileheader,
+    read_traceheader,
+    read_file,
+    segy_read,
+)
+from .scan import SegyScan, segy_scan, save_scan, load_scan
+from .write import (
+    write_fileheader,
+    write_traceheader,
+    write_block,
+    segy_write,
+)
+from .utils import get_header
+
+__all__ = [
+    "BinaryFileHeader",
+    "BinaryTraceHeader",
+    "FileHeader",
+    "SeisBlock",
+    "SegyScan",
+    "read_fileheader",
+    "read_traceheader",
+    "read_file",
+    "segy_read",
+    "segy_scan",
+    "save_scan",
+    "load_scan",
+    "write_fileheader",
+    "write_traceheader",
+    "write_block",
+    "segy_write",
+    "get_header",
+]
+
+logging.basicConfig(level=logging.INFO)

pysegy/ibm.py

@@ -0,0 +1,67 @@
+"""
+Conversion helpers between IBM and IEEE floating point formats.
+"""
+
+from typing import Union
+
+
+def ibm_to_ieee(value: Union[bytes, bytearray, int]) -> float:
+    """
+    Convert a 4-byte IBM floating point number to ``float``.
+
+    Parameters
+    ----------
+    value : bytes or int
+        Input value encoded in the IBM 32-bit floating point format.
+
+    Returns
+    -------
+    float
+        Floating point number in IEEE representation.
+    """
+    if isinstance(value, (bytes, bytearray)):
+        if len(value) != 4:
+            raise ValueError("IBM float must be 4 bytes")
+        val = int.from_bytes(value, byteorder='big', signed=False)
+    else:
+        val = value & 0xffffffff
+    if val == 0:
+        return 0.0
+    sign = 1 if (val >> 31) == 0 else -1
+    exponent = (val >> 24) & 0x7f
+    fraction = val & 0x00ffffff
+    # IBM exponent is base 16 biased by 64
+    mant = fraction / float(0x01000000)
+    return sign * mant * 16 ** (exponent - 64)
+
+
+def ieee_to_ibm(f: float) -> bytes:
+    """
+    Convert ``float`` to IBM 32-bit floating point bytes.
+
+    Parameters
+    ----------
+    f : float
+        Numeric value in IEEE representation.
+
+    Returns
+    -------
+    bytes
+        The value encoded using the IBM format.
+    """
+    if f == 0.0:
+        return b"\x00\x00\x00\x00"
+    sign = 0
+    if f < 0:
+        sign = 0x80
+        f = -f
+    exponent = 64
+    while f < 1.0:
+        f *= 16.0
+        exponent -= 1
+    while f >= 16.0:
+        f /= 16.0
+        exponent += 1
+    fraction = int(f * 0x01000000) & 0x00ffffff
+    val = (sign << 24) | (exponent << 24) | fraction
+    return val.to_bytes(4, byteorder='big')