malac-utils 1.3.3__tar.gz → 1.4.1a74__tar.gz

{malac_utils-1.3.3/malac_utils.egg-info → malac_utils-1.4.1a74}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: malac-utils
-Version: 1.3.3
+Version: 1.4.1a74
 Summary: Mapping Language Compiler Utils
 Author-email: cdeHealth-Team <contact-project+cdehealth-malac-hd-52276676-issue-@incoming.gitlab.com>
 Project-URL: Homepage, https://gitlab.com/cdehealth/malac-hd
@@ -23,7 +23,8 @@ Classifier: Topic :: Software Development :: Libraries
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: python-dateutil>=2.8.2
+Provides-Extra: test
+Requires-Dist: pytest<10,>=9; extra == "test"
 Dynamic: license-file
 
 # MaLaC Utils
@@ -32,17 +33,13 @@ MaLaC-HD (MApping LAnguage Compiler for Health Data) is a tool that you can use
 
 This is the utils package to execute generated mappings. You can use it independently from the `malac-hd` package, together with the appropriate model packages.
 
-[TOC]
+It contains the Python modules mappings compiled to Python use (`malac.utils`), and ships the runtimes of mappings compiled to JavaScript, Java and C#, which malac-hd runs them with:
 
-## Contributing and Support
+- `malac/utils/js/malac-utils.js`, the JavaScript runtime as one self-contained file for Node.js and web browsers (also published to npm as [`malac-utils`](https://www.npmjs.com/package/malac-utils)),
+- `malac/utils/java/malac-utils.jar`, the Java runtime as one self-contained jar (also published to Maven Central as `io.gitlab.cdehealth:malac-utils`),
+- `malac/utils/dotnet/malac-utils.dll`, the .NET runtime (also published to NuGet as [`Malac.Utils`](https://www.nuget.org/packages/Malac.Utils)).
 
-Please read [CONTRIBUTING.md](CONTRIBUTING.md) for contributing to cdeHealth projects which are hosted in the [cdeHealth group](https://gitlab.com/cdehealth) on GitLab.com.
-Please read [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md) to make participation in our community a harassment free experience for everyone.
-
-## Authors and acknowledgment
-We want to thank
-- [ELGA GmbH](https://www.elga.gv.at/) with their [CDA2FHIR](https://collab.hl7.at/display/BAL/AG+ELGA+CDA+Laborbefund+zu+FHIR) projects and
-- [AIT Austrian Institute of Technology GmbH](https://www.ait.ac.at/) with their [SmartFOX](https://www.smart-fox.at/) project.
+See the [repository](https://gitlab.com/cdehealth/malac/utils) for their sources and how to build them.
 
 ## License
-This is a LGPL licensed project.
+This is a LGPL licensed project (LGPL-3.0-only).

malac_utils-1.4.1a74/README.md

@@ -0,0 +1,16 @@
+# MaLaC Utils
+
+MaLaC-HD (MApping LAnguage Compiler for Health Data) is a tool that you can use to convert mappings between different health data formats to executable code. It can also be used as a library to dynamically execute mappings.
+
+This is the utils package to execute generated mappings. You can use it independently from the `malac-hd` package, together with the appropriate model packages.
+
+It contains the Python modules mappings compiled to Python use (`malac.utils`), and ships the runtimes of mappings compiled to JavaScript, Java and C#, which malac-hd runs them with:
+
+- `malac/utils/js/malac-utils.js`, the JavaScript runtime as one self-contained file for Node.js and web browsers (also published to npm as [`malac-utils`](https://www.npmjs.com/package/malac-utils)),
+- `malac/utils/java/malac-utils.jar`, the Java runtime as one self-contained jar (also published to Maven Central as `io.gitlab.cdehealth:malac-utils`),
+- `malac/utils/dotnet/malac-utils.dll`, the .NET runtime (also published to NuGet as [`Malac.Utils`](https://www.nuget.org/packages/Malac.Utils)).
+
+See the [repository](https://gitlab.com/cdehealth/malac/utils) for their sources and how to build them.
+
+## License
+This is a LGPL licensed project (LGPL-3.0-only).

malac_utils-1.4.1a74/malac/utils/__init__.py

@@ -0,0 +1,2 @@
+
+__version__ = "1.4.1-alpha.74"

malac_utils-1.4.1a74/malac/utils/date.py

@@ -0,0 +1,61 @@
+import re
+import datetime
+
+# FHIR date, dateTime and instant as well as FHIRPath date and dateTime literals: YYYY, YYYY-MM
+# or YYYY-MM-DD, followed by T and hh, hh:mm or hh:mm:ss with optional fractional seconds and an
+# offset Z or +/-hh:mm up to 14:00, see https://hl7.org/fhir/datatypes.html#primitive and
+# https://hl7.org/fhirpath/#datetime (also a space as separator, as str(datetime) gives)
+FHIR_PATTERN = re.compile(r"^([0-9]{4})(?:-([0-9]{2})(?:-([0-9]{2}))?)?(?:[T ]([0-9]{2})(?::([0-9]{2})(?::([0-9]{2})(?:\.([0-9]+))?)?)?(Z|[+-][0-9]{2}:?[0-9]{2})?|T)?$")
+
+# CDA TS: any portion from YYYY to YYYYMMDDhhmmss, fractional seconds only if precise
+# to the second, and an optional offset +/-HHMM, see
+# https://build.fhir.org/ig/HL7/CDA-core-2.0/StructureDefinition-ts-simple.html
+TS_PATTERN = re.compile(r"^([0-9]{4})(?:([0-9]{2})(?:([0-9]{2})(?:([0-9]{2})(?:([0-9]{2})(?:([0-9]{2})(?:\.([0-9]+))?)?)?)?)?)?([+-][0-9]{1,4})?$")
+
+# FHIR permits offsets from -14:00 to +14:00, which also bounds CDA offsets mapped to FHIR
+MAX_OFFSET = datetime.timedelta(hours=14)
+
+class PartialDate:
+
+    def __init__(self, value):
+        self.value = value
+
+    def isoformat(self):
+        return self.value
+
+    def __repr__(self):
+        return self.value
+
+def _tzinfo(offset):
+    if not offset:
+        return None
+    if offset == "Z":
+        return datetime.timezone.utc
+    digits = offset[1:].replace(":", "")
+    hours, minutes = (digits, 0) if len(digits) <= 2 else (digits[:-2], digits[-2:])
+    delta = datetime.timedelta(hours=int(hours), minutes=int(minutes))
+    if int(minutes) > 59 or delta > MAX_OFFSET:
+        raise ValueError("Invalid offset: %s" % offset)
+    return datetime.timezone(-delta if offset[0] == "-" else delta)
+
+def _datetime(year, month, day, hour, minute, second, fraction, offset):
+    tzinfo = _tzinfo(offset)
+    microsecond = int(fraction[:6].ljust(6, "0")) if fraction else 0 # datetime supports microseconds only
+    return datetime.datetime(int(year), int(month), int(day), int(hour or 0), int(minute or 0), int(second or 0), microsecond, tzinfo)
+
+def parse(datestr):
+    datestr = getattr(datestr, "value", datestr).strip()
+    if (match := FHIR_PATTERN.match(datestr)) and (match.group(3) or not match.group(4)): # time only with a full date
+        year, month, day = match.groups()[:3]
+        if not day:
+            return PartialDate("-".join(part for part in (year, month) if part))
+        return _datetime(*match.groups())
+    elif match := TS_PATTERN.match(datestr):
+        year, month, day, hour = match.groups()[:4]
+        if not hour:
+            # date precision, an offset is not representable here but must still be valid
+            _tzinfo(match.group(8))
+            return PartialDate("-".join(part for part in (year, month, day) if part))
+        return _datetime(*match.groups())
+    else:
+        raise ValueError("Unknown date format: %s" % datestr)

malac_utils-1.4.1a74/malac/utils/exceptions.py

@@ -0,0 +1,10 @@
+"""
+Exception hierarchy for MaLaC-HD-generated code
+"""
+
+class MappingExecutionError(Exception):
+    """Base class for errors raised while executing a mapping."""
+
+    
+class CheckFailedError(MappingExecutionError):
+    """Error in case a check clause failed."""

{malac_utils-1.3.3 → malac_utils-1.4.1a74}/malac/utils/fhirpath.py

@@ -7,6 +7,7 @@ import math
 from urllib.parse import urlparse
 
 from . import date as dateutil
+from . import html
 
 time_units = ["year", "years", "month", "months", "week", "weeks", "day", "days", "hour", "hours", "minute", "minutes", "second", "seconds", "millisecond", "milliseconds"]
 
@@ -100,6 +101,8 @@ class FHIRPathUtils(object):
         }
 
     def get(self, val, *attrs, **options):
+        if len(attrs) == 1 and options.get("strip", False):
+            return [html.strip_tags(val)]
         attr = attrs[0]
         try:
             result = getattr(val, attr) or []
@@ -119,10 +122,6 @@ class FHIRPathUtils(object):
                         prefix = getattr(val, key)
                         if prefix and attr.lower() == (prefix + key[:-10]).lower():
                             result = getattr(val, key[:-10]) or []
-        if result != [] and len(attrs) == 1 and options.get("strip", False):
-            result = str(result).strip()
-            if result == "":
-                result = []
         result = result if isinstance(result, list) else [result]
         if len(attrs) == 1:
             return result
@@ -369,7 +368,12 @@ class FHIRPathUtils(object):
             unit = val_a if isinstance(val_a, self.model.Quantity) else val_b # TODO divide units (g / m -> g/m)
             return [self.model.Quantity(value=self.model.decimal(value=v_a / v_b), unit=unit.unit, code=unit.code, system=unit.system)]
         else:
-            return [decimal.Decimal(round(val_a / val_b, 8))]
+            # decimal division, also for integers (not float division)
+            if isinstance(val_a, int):
+                val_a = decimal.Decimal(val_a)
+            if isinstance(val_b, int):
+                val_b = decimal.Decimal(val_b)
+            return [round(val_a / val_b, 8)]
 
     def div(self, val_a, val_b):
         val_a = self.single_value(val_a)
@@ -563,6 +567,14 @@ class FHIRPathUtils(object):
         else:
             return [-1]
 
+    def lastIndexOf(self, val, other):
+        val = self.string_value(val)
+        other = self.string_value(self.single_value(other, str, self.model.string))
+        if val == [] or other == []:
+            return []
+        else:
+            return [val.rfind(other)]
+
     def concat(self, val, other):
         val = self.string_value(self.single_value(val))
         other = self.string_value(self.single_value(other))
@@ -1106,13 +1118,13 @@ class FHIRPathUtils(object):
             return [val]
         elif isinstance(val, self.model.decimal):
             return [decimal.Decimal(val.value)]
+        elif isinstance(val, self.model.boolean) or isinstance(val, bool): # before int, which bool is a subclass of
+            return [decimal.Decimal("1.0") if self.boolean_value(val) else decimal.Decimal("0.0")]
         elif isinstance(val, self.model.integer) or isinstance(val, int):
             if isinstance(val, self.model.integer):
                 return [decimal.Decimal(val.value)]
             else:
                 return [decimal.Decimal(val)]
-        elif isinstance(val, self.model.boolean) or isinstance(val, bool):
-            return [decimal.Decimal("1.0") if self.boolean_value(val) else decimal.Decimal("0.0")]
         elif isinstance(val, self.model.string) or isinstance(val, str):
             try:
                 return [decimal.Decimal(self.string_value(val))]

malac_utils-1.4.1a74/malac/utils/html.py

@@ -0,0 +1,176 @@
+import io
+import logging
+from html import escape as html_escape
+from html.parser import HTMLParser
+
+logger = logging.getLogger(__name__)
+
+def _attr(name, value):
+    return f' {name}="{html_escape(str(value))}"'
+
+def _handle_tr(tr):
+    html = ""
+    attrs = ""
+    if tr.ID:
+        attrs += _attr("id", tr.ID)
+    if tr.styleCode:
+        attrs += _attr("class", tr.styleCode)
+    html += "<tr %s>" % attrs
+    for th in tr.th:
+        html += _handle_tdh(th, "th")
+    for td in tr.td:
+        html += _handle_tdh(td, "td")
+    html += "</tr>"
+    return html
+
+def _handle_tdh(tdh, tag):
+    html = ""
+    attrs = ""
+    if tdh.ID:
+        attrs += _attr("id", tdh.ID)
+    if tdh.styleCode:
+        attrs += _attr("class", tdh.styleCode)
+    html += f"<{tag}%s>" % attrs
+    html += _handle_content(tdh)
+    html += f"</{tag}>"
+    return html
+
+def _handle_content(elem):
+    html = ""
+    for content in elem.content_:
+        if content.category == 1:
+            html += html_escape(content.value)
+        elif content.category == 2:
+            if content.name == "br":
+                html += f'<br />'
+            else:
+                logger.warning("Skipping unhandled simple narrative content '%s'", content.name)
+        elif content.category == 3:
+            attrs = ""
+            if getattr(content.value, "ID", None):
+                attrs += _attr("id", content.value.ID)
+            if content.name == "table":
+                html += "<table%s>" % attrs
+                if content.value.thead:
+                    html += "<thead>"
+                    for tr in content.value.thead.tr:
+                        html += _handle_tr(tr)
+                    html += "</thead>"
+                if content.value.tfoot:
+                    html += "<tfoot>"
+                    for tr in content.value.tfoot.tr:
+                        html += _handle_tr(tr)
+                    html += "</tfoot>"
+                for tbody in content.value.tbody or [None]:
+                    html += "<tbody>"
+                    for tr in (tbody.tr if tbody else []):
+                        html += _handle_tr(tr)
+                    html += "</tbody>"
+                html += "</table>"
+            elif content.name == "content":
+                if content.value.styleCode == "Italics":
+                    attrs += ' class="italics"'
+                elif content.value.styleCode:
+                    attrs += _attr("class", content.value.styleCode)
+                html += "<span%s>" % attrs
+                html += _handle_content(content.value)
+                html += "</span>"
+            elif content.name == "paragraph":
+                if content.value.styleCode:
+                    attrs += _attr("class", content.value.styleCode)
+                html += "<p%s>" % attrs
+                html += _handle_content(content.value)
+                html += "</p>"
+            elif content.name == "sup":
+                html += "<sup%s>" % attrs
+                html += _handle_content(content.value)
+                html += "</sup>"
+            elif content.name == "footnote":
+                html += '<span class="footnote"%s>' % attrs
+                html += _handle_content(content.value)
+                html += "</span>"
+            elif content.name == "renderMultiMedia":
+                html += '<img%s%s />' % (_attr("src", content.value.referencedObject), attrs)
+            elif content.name == "br":
+                html += f'<br />'
+            elif content.name == "caption":
+                html += "<caption%s>" % attrs
+                html += _handle_content(content.value)
+                html += "</caption>"
+            elif content.name == "list":
+                if content.value.listType in [None, "unordered"]:
+                    if content.value.styleCode:
+                        attrs += _attr("style", f"list-style-type:{content.value.styleCode.lower()}")
+                    html += "<ul%s>" % attrs
+                    for item in content.value.item:
+                        html += "<li>"
+                        html += _handle_content(item)
+                        html += "</li>"
+                    html += "</ul>"
+                elif content.value.listType == "ordered":
+                    styleCodeTrans = {"LittleRoman": "i", "BigRoman": "I", "LittleAlpha": "a", "BigAlpha": "A", "Arabic": "1"}
+                    listType = styleCodeTrans.get(content.value.styleCode)
+                    if listType:
+                        attrs += f' type="{listType}"'
+                    html += "<ol%s>" % attrs
+                    for item in content.value.item:
+                        html += "<li>"
+                        html += _handle_content(item)
+                        html += "</li>"
+                    html += "</ol>"
+            else:
+                html += _handle_content(content.value)
+        else:
+            logger.warning("Skipping unhandled narrative content category %s", content.category)
+    return html
+
+def strucdoctext2html(module, strucdoctext):
+    html = _handle_content(strucdoctext)
+    div = module.Flow(content_=[module.MixedContainer(1, 0, "", html)])
+    div.original_tagname_ = "div"
+    div.valueOf_ = ""
+    return div
+
+def ed2html(module, ed):
+    # TODO: improve
+    div = module.Flow(content_=[module.MixedContainer(1, 0, "", f"<v3:reference>{ed.reference.value}</v3:reference>")])
+    div.original_tagname_ = "div"
+    div.valueOf_ = ""
+    return div
+
+def ed2markdown(module, ed):
+    # TODO: improve
+    return module.markdown(value=f"<v3:reference>{ed.reference.value}</v3:reference>")
+
+def strucdoctext2markdown(module, strucdoctext):
+    # TODO: implement properly
+    return module.markdown(value=strip_tags(strucdoctext))
+
+
+def strip_tags(val):
+    class MLStripper(HTMLParser):
+        def __init__(self):
+            super().__init__()
+            self.reset()
+            self.strict = False
+            self.convert_charrefs= True
+            self.text = io.StringIO()
+        def handle_data(self, d):
+            self.text.write(d)
+            self.text.write(" ")
+        def get_data(self):
+            return self.text.getvalue()
+
+    html = _handle_content(val)
+    s = MLStripper()
+    s.feed(html)
+    text = s.get_data()
+    text = " ".join(text.split())
+    return text
+
+
+def builddiv(module, content):
+    div = module.Flow(content_=[module.MixedContainer(1, 0, "", content)])
+    div.original_tagname_ = "div"
+    div.valueOf_ = ""
+    return div