gismap 0.1.0__py3-none-any.whl → 0.2.1__py3-none-any.whl

gismap/lab/vis.py

@@ -4,48 +4,154 @@ import json
 
 
 vis_template = Template("""
+<div class="gismap-content">
 <div id="${container_id}"></div>
+<a
+  href="https://balouf.github.io/gismap/"
+  target="_blank"
+  id="gismap-brand"
+  style="position: absolute; left: 10px; bottom: 10px; text-decoration: none; color: #888; font-size: min(2vw, 10px);
+  z-index: 10; pointer-events: auto;"
+>
+  &copy; Gismap 2025
+</a>
 <div id="${modal_id}" class="modal">
-  <div class="modal-content">
+<div class="modal-content">
     <span class="close" id="${modal_close_id}">&times;</span>
     <div id="${modal_body_id}"></div>
   </div>
 </div>
+</div>
 <style>
+.gismap-content {
+position: relative;
+width: 100%;
+height: 80vh !important;
+max-width: 100vw;
+max-height: 100vh !important;
+}
   /* Styles adaptatifs pour dark/light */
+/* Default: dark mode styles */
+#${container_id} {
+  width: 100%;
+  height: 100%;
+  box-sizing: border-box;
+  border: 1px solid #444;
+  background: #181818;
+}
+
+.modal {
+  display: none;
+  position: fixed;
+  z-index: 1000;
+  left: 0; top: 0;
+  width: 100%; height: 100%;
+  overflow: auto;
+  background-color: rgba(10,10,10,0.85);
+}
+
+.modal-content {
+  background-color: #23272e;
+  color: #f0f0f0;
+  margin: 10% auto;
+  padding: 24px;
+  border: 1px solid #888;
+  width: 50%;
+  border-radius: 8px;
+  box-shadow: 0 5px 15px rgba(0,0,0,.6);
+}
+
+.close {
+  color: #aaa;
+  float: right;
+  font-size: 28px;
+  font-weight: bold;
+  cursor: pointer;
+}
+
+.close:hover, .close:focus {
+  color: #fff;
+  text-decoration: none;
+  cursor: pointer;
+}
+
+/* PyData Sphinx Light Theme */
+html[data-theme="light"] #${container_id},
+body[data-jp-theme-light="true"] #${container_id} {
+  background: #fff;
+  border: 1px solid #ccc;
+}
+
+html[data-theme="light"] .modal,
+body[data-jp-theme-light="true"] .modal {
+  background-color: rgba(220,220,220,0.85);
+}
+
+html[data-theme="light"] .modal-content,
+body[data-jp-theme-light="true"] .modal-content {
+  background: #fff;
+  color: #222;
+  border: 1px solid #888;
+}
+
+html[data-theme="light"] .close,
+body[data-jp-theme-light="true"] .close {
+  color: #222;
+}
+
+html[data-theme="light"] .close:hover, html[data-theme="light"] .close:focus,
+body[data-jp-theme-light="true"] .close:hover, body[data-jp-theme-light="true"] .close:focus {
+  color: #555;
+}
+
+/* Fallback: system light mode */
+@media (prefers-color-scheme: light) {
   #${container_id} {
-    width: 1200px; height: 800px; border: 1px solid #444;
-    background: #181818;
+    background: #fff;
+    border: 1px solid #ccc;
   }
   .modal {
-    display: none; position: fixed; z-index: 1000; left: 0; top: 0; width: 100%; height: 100%;
-    overflow: auto; background-color: rgba(10,10,10,0.85);
+    background-color: rgba(220,220,220,0.85);
   }
   .modal-content {
-    background-color: #23272e; color: #f0f0f0;
-    margin: 10% auto; padding: 24px; border: 1px solid #888;
-    width: 50%; border-radius: 8px;
-    box-shadow: 0 5px 15px rgba(0,0,0,.6);
+    background: #fff;
+    color: #222;
+    border: 1px solid #888;
   }
   .close {
-    color: #aaa; float: right; font-size: 28px; font-weight: bold; cursor: pointer;
+    color: #222;
   }
-  .close:hover, .close:focus { color: #fff; text-decoration: none; cursor: pointer; }
-  @media (prefers-color-scheme: light) {
-    #${container_id} { background: #fff; border: 1px solid #ccc; }
-    .modal { background-color: rgba(220,220,220,0.85); }
-    .modal-content { background: #fff; color: #222; border: 1px solid #888; }
-    .close { color: #222; }
-    .close:hover, .close:focus { color: #555; }
+  .close:hover, .close:focus {
+    color: #555;
   }
+}
 </style>
-<script type="text/javascript" src="https://unpkg.com/vis-network/standalone/umd/vis-network.min.js"></script>
 <script type="text/javascript">
 (function() {
   // Détection du thème
-  function getTheme() {
-    return window.matchMedia('(prefers-color-scheme: dark)').matches ? 'dark' : 'light';
+function getTheme() {
+  // Try PyData Sphinx theme on <html>
+  const pydataTheme = document.documentElement.getAttribute("data-theme");
+  if (pydataTheme === "dark" || pydataTheme === "light") {
+    return pydataTheme;
   }
+
+  // Try JupyterLab theme on <body>
+  const jupyterLabTheme = document.body.getAttribute("data-jp-theme-name");
+  if (jupyterLabTheme) {
+    // Simplify theme name to 'dark' or 'light'
+    const lowerName = jupyterLabTheme.toLowerCase();
+    if (lowerName.includes("dark")) {
+      return "dark";
+    }
+    if (lowerName.includes("light")) {
+      return "light";
+    }
+  }
+
+  // Fallback to system preference
+  return window.matchMedia && window.matchMedia('(prefers-color-scheme: dark)').matches ? "dark" : "light";
+};
   function getVisOptions(theme) {
     if (theme === 'dark') {
       return {
@@ -86,15 +192,41 @@ vis_template = Template("""
         interaction: { hover: true }
       };
     }
+  };
+
+  var physics = {
+  physics: {
+    solver: "forceAtlas2Based",
+    forceAtlas2Based: {
+      gravitationalConstant: -50,
+      centralGravity: 0.01,
+      springLength: 200,
+      springConstant: 0.08,
+      damping: 0.98,
+      avoidOverlap: 1
+    },
+    maxVelocity: 10,
+    minVelocity: 0.9,
+    stabilization: {
+      enabled: true,
+      iterations: 2000,
+      updateInterval: 50,
+      onlyDynamicEdges: false,
+      fit: true
+    },
+    timestep: 0.25
   }
+};
+
+  function renderNetwork() {
   const nodes = new vis.DataSet(${nodes_json});
   const edges = new vis.DataSet(${edges_json});
   const container = document.getElementById('${container_id}');
   let network = null;
-  function renderNetwork() {
     const theme = getTheme();
     const options = getVisOptions(theme);
     network = new vis.Network(container, { nodes: nodes, edges: edges }, options);
+    network.setOptions(physics)
     // Tooltip survol
     network.on("hoverNode", function(params) {
       const node = nodes.get(params.node);
@@ -131,12 +263,38 @@ vis_template = Template("""
     window.onclick = function(event) {
       if (event.target == modal) { modal.style.display = "none"; }
     };
+  };
+
+  function loadVisAndRender() {
+  if (typeof vis === 'undefined') {
+    var script = document.createElement('script');
+    script.src = "https://unpkg.com/vis-network/standalone/umd/vis-network.min.js";
+    script.type = "text/javascript";
+    script.onload = function() {
+      console.log("vis-network loaded dynamically");
+      renderNetwork();  // Graph init after vis is loaded
+    };
+    document.head.appendChild(script);
+  } else {
+    console.log("vis-network already loaded");
+    renderNetwork();  // Graph init immediately
   }
-  renderNetwork();
+}
+loadVisAndRender();
+
   // Adapter dynamiquement si le thème change
-  window.matchMedia('(prefers-color-scheme: dark)').addEventListener('change', function() {
-    renderNetwork();
-  });
+  window.addEventListener("theme-changed", () => loadVisAndRender());
+  const observer = new MutationObserver(mutations => {
+  for (const mutation of mutations) {
+    if (mutation.type === "attributes" && mutation.attributeName === "data-jp-theme-name") {
+      loadVisAndRender();
+    }
+  }
+    });
+    observer.observe(document.body, { attributes: true });
+    if (window.matchMedia) {
+      window.matchMedia('(prefers-color-scheme: dark)').addEventListener('change', () => loadVisAndRender());
+    };
 })();
 </script>
 """)

gismap/sources/dblp.py

@@ -5,15 +5,15 @@ from bs4 import BeautifulSoup as Soup
 from time import sleep
 
 from gismap.sources.models import DB, Author, Publication
-from gismap.utils.text import clean_aliases
+from gismap.utils.text import clean_aliases, auto_int
 from gismap.utils.requests import get
 
 
 @dataclass(repr=False)
 class DBLP(DB):
     db_name: ClassVar[str] = "dblp"
-    author_backoff: ClassVar[float] = 7.0
-    publi_backoff: ClassVar[float] = 2.0
+    author_backoff: ClassVar[float] = 5.0
+    publi_backoff: ClassVar[float] = 1.0
 
     @classmethod
     def search_author(cls, name, wait=True):
@@ -84,17 +84,16 @@ class DBLP(DB):
         authors=[DBLPAuthor(name='Yacine Boufkhad', key='75/5742'), DBLPAuthor(name='Fabien Mathieu', key='66/2077'),
         DBLPAuthor(name='Fabien de Montgolfier', key='57/6313'), DBLPAuthor(name='Diego Perino', key='03/3645'),
         DBLPAuthor(name='Laurent Viennot', key='v/LaurentViennot')],
-        venue='IPTPS', type='conference', year=2008, key='conf/iptps/BoufkhadMMPV08',
-        url='https://dblp.org/rec/conf/iptps/BoufkhadMMPV08.html', pages=4)
+        venue='IPTPS', type='conference', year=2008, key='conf/iptps/BoufkhadMMPV08')
         >>> publications[-1] # doctest:  +NORMALIZE_WHITESPACE
         DBLPPublication(title='Upper Bounds for Stabilization in Acyclic Preference-Based Systems.',
         authors=[DBLPAuthor(name='Fabien Mathieu', key='66/2077')], venue='SSS', type='conference', year=2007,
-        key='conf/sss/Mathieu07', url='https://dblp.org/rec/conf/sss/Mathieu07.html', pages='372-382')
+        key='conf/sss/Mathieu07')
         """
         r = get(f"https://dblp.org/pid/{a.key}.xml")
         soup = Soup(r, features="xml")
         if wait:
-            sleep(cls.author_backoff)
+            sleep(cls.publi_backoff)
         res = [DBLPPublication.from_soup(r) for r in soup("r")]
         return [p for p in res if p.authors]
 
@@ -110,7 +109,7 @@ class DBLPAuthor(Author, DBLP):
             return f"https://dblp.org/pid/{self.key}.html"
         return f"https://dblp.org/search?q={quote_plus(self.name)}"
 
-    def get_publications(self, wait=False):
+    def get_publications(self, wait=True):
         return DBLP.from_author(self, wait=wait)
 
 
@@ -128,29 +127,27 @@ DBLP_TYPES = {
 @dataclass(repr=False)
 class DBLPPublication(Publication, DBLP):
     key: str
-    url: str = None
-    pages: str = None
-    volume: int = None
-    number: int = None
+    metadata: dict = field(default_factory=dict)
+
+    @property
+    def url(self):
+        if self.key:
+            return f"https://dblp.org/rec/{self.key}.html"
+        else:
+            return None
 
     @classmethod
     def from_soup(cls, soup):
         p = soup.find()
         typ = p.get("publtype", p.name)
         typ = DBLP_TYPES.get(typ, typ)
+
         res = {
             "type": typ,
             "key": p["key"],
-            "url": f"https://dblp.org/rec/{p['key']}.html",
+            "title": p.title.text,
+            "year": int(p.year.text),
         }
-        keys = ["title", "booktitle", "pages", "journal", "year", "volume", "number"]
-        for tag in keys:
-            t = p.find(tag)
-            if t:
-                try:
-                    res[tag] = int(t.text)
-                except ValueError:
-                    res[tag] = t.text
         for tag in ["booktitle", "journal"]:
             t = p.find(tag)
             if t:
@@ -159,4 +156,11 @@ class DBLPPublication(Publication, DBLP):
         else:
             res["venue"] = "unpublished"
         res["authors"] = [DBLPAuthor(key=a["pid"], name=a.text) for a in p("author")]
-        return cls(**{k: v for k, v in res.items() if k in cls.__match_args__})
+
+        metadata = dict()
+        for tag in p.find_all(recursive=False):
+            name = tag.name
+            if name not in {"title", "year", "author", "booktitle", "journal"}:
+                metadata[name] = auto_int(tag.text)
+
+        return cls(**res, metadata=metadata)

gismap/sources/hal.py

@@ -2,6 +2,7 @@ from typing import ClassVar
 from dataclasses import dataclass, field
 from collections import defaultdict
 from urllib.parse import quote_plus
+from bs4 import BeautifulSoup as Soup
 import json
 
 from gismap.sources.models import DB, Publication, Author  #  DBAuthor, DBPublication
@@ -67,7 +68,7 @@ class HAL(DB):
                 elif "person_i" in a:
                     pids[a["person_i"]].add(a.get("label_s"))
             elif "fullName_s" in a:
-                    names.add(a["fullName_s"])
+                names.add(a["fullName_s"])
         res = [
             HALAuthor(name=name, key=k, aliases=clean_aliases(name, v))
             for k, v in hids.items()
@@ -84,7 +85,7 @@ class HAL(DB):
                 HALAuthor(
                     name=name,
                     key=fullname,
-                    aliases=clean_aliases(name, fullname),
+                    aliases=[],
                     key_type="fullname",
                 )
                 for fullname in names
@@ -117,7 +118,7 @@ class HAL(DB):
         HALAuthor(name='Diego Perino', key='Diego Perino', key_type='fullname'),
         HALAuthor(name='Laurent Viennot', key='laurentviennot')],
         venue='Proceedings of the 7th Internnational Workshop on Peer-to-Peer Systems (IPTPS)', type='conference',
-        year=2008, key='471724', url='https://inria.hal.science/inria-00471724v1')
+        year=2008, key='471724')
         >>> diego = publications[2].authors[3]
         >>> diego
         HALAuthor(name='Diego Perino', key='Diego Perino', key_type='fullname')
@@ -127,7 +128,7 @@ class HAL(DB):
         HALPublication(title='Upper bounds for stabilization in acyclic preference-based systems',
         authors=[HALAuthor(name='Fabien Mathieu', key='fabien-mathieu')],
         venue="SSS'07 - 9th international conference on Stabilization, Safety, and Security of Distributed Systems",
-        type='conference', year=2007, key='668356', url='https://inria.hal.science/hal-00668356v1')
+        type='conference', year=2007, key='668356')
 
         Case of someone with multiple ids one want to cumulate:
 
@@ -180,9 +181,30 @@ class HALAuthor(Author, HAL):
     key: str | int = None
     key_type: str = None
     aliases: list = field(default_factory=list)
+    _url: str = None
+    _img: str = None
+    _cv: bool = None
+
+    def check_cv(self):
+        if self.key_type is not None:
+            self._cv = False
+            return None
+        url = f"https://cv.hal.science/{self.key}"
+        soup = Soup(get(url), "lxml")
+        if soup.form:
+            self._cv = False
+            return None
+        self._cv = True
+        self._url = url
+        try:
+            self._img = soup.main.section.div.div.div.img["src"]
+        except TypeError:
+            return None
 
     @property
     def url(self):
+        if self._url is not None:
+            return self._url
         if self.key_type == "pid":
             return f"https://hal.science/search/index/?q=*&authIdPerson_i={self.key}"
         elif self.key_type == "fullname":
@@ -190,6 +212,12 @@ class HALAuthor(Author, HAL):
         else:
             return f"https://hal.science/search/index/?q=*&authIdHal_s={self.key}"
 
+    @property
+    def img(self):
+        if self._cv is None:
+            self.check_cv()
+        return self._img
+
     def get_publications(self):
         return HAL.from_author(self)
 
@@ -241,8 +269,11 @@ HAL_KEYS = {
 @dataclass(repr=False)
 class HALPublication(Publication, HAL):
     key: str
-    abstract: str = None
-    url: str = None
+    metadata: dict = field(default_factory=dict)
+
+    @property
+    def url(self):
+        return self.metadata.get("url")
 
     @classmethod
     def from_json(cls, r):
@@ -258,15 +289,20 @@ class HALPublication(Publication, HAL):
         :class:`~gismap.sources.hal.HALPublication`
 
         """
-        res = {v: unlist(r[k]) for k, v in HAL_KEYS.items() if k in r}
+        keys = {v: unlist(r[k]) for k, v in HAL_KEYS.items() if k in r}
+        res = {k: keys[k] for k in ["key", "title", "year"]}
+        # res = {v: unlist(r[k]) for k, v in HAL_KEYS.items() if k in r}
         res["authors"] = [
             parse_facet_author(a) for a in r.get("authFullNamePersonIDIDHal_fs", [])
         ]
         for tag in ["booktitle", "journal", "conference"]:
-            if tag in res:
-                res["venue"] = res[tag]
+            if tag in keys:
+                res["venue"] = keys[tag]
                 break
         else:
             res["venue"] = "unpublished"
-        res["type"] = HAL_TYPES.get(res["type"], res["type"].lower())
-        return cls(**{k: v for k, v in res.items() if k in cls.__match_args__})
+        res["type"] = HAL_TYPES.get(keys["type"], keys["type"].lower())
+        res["metadata"] = {
+            k: keys[k] for k in {"abstract", "url"} if k in keys and keys[k]
+        }
+        return cls(**res)

gismap/sources/multi.py

@@ -6,11 +6,22 @@ from gismap.sources.models import Publication, Author
 from gismap.utils.text import clean_aliases
 
 
-score_rosetta = {
-    "db_name": {"dblp": 1, "hal": 2},
-    "venue": {"CoRR": -1, "unpublished": -2},
-    "type": {"conference": 1, "journal": 2},
-}
+def score_author_source(dbauthor):
+    if dbauthor.db_name == "hal":
+        if dbauthor.key_type == "fullname":
+            return -1
+        elif dbauthor.key_type == "pid":
+            return 2
+        else:
+            return 3
+    elif dbauthor.db_name == "dblp":
+        return 1
+    else:
+        return 0
+
+
+def sort_author_sources(sources):
+    return sorted(sources, key=score_author_source, reverse=True)
 
 
 @dataclass(repr=False)
@@ -35,10 +46,15 @@ class SourcedAuthor(Author):
 
     @classmethod
     def from_sources(cls, sources):
+        sources = sort_author_sources(sources)
         return cls(name=sources[0].name, sources=sources)
 
-    def get_publications(self, clean=True):
-        res = {p.key: p for a in self.sources for p in a.get_publications()}
+    def get_publications(self, clean=True, selector=None):
+        if selector is None:
+            selector = []
+        res = {
+            p.key: p for a in self.sources for p in a.get_publications() if all(f(p) for f in selector)
+        }
         if clean:
             regroup_authors({self.key: self}, res)
             return regroup_publications(res)
@@ -46,30 +62,49 @@ class SourcedAuthor(Author):
             return res
 
 
+publication_score_rosetta = {
+    "db_name": {"dblp": 1, "hal": 2},
+    "venue": {"CoRR": -1, "unpublished": -2},
+    "type": {"conference": 1, "journal": 2},
+}
+
+
+def score_publication_source(source):
+    scores = [
+        v.get(getattr(source, k, None), 0) for k, v in publication_score_rosetta.items()
+    ]
+    scores.append(source.year)
+    return tuple(scores)
+
+
+def sort_publication_sources(sources):
+    return sorted(sources, key=score_publication_source, reverse=True)
+
+
 @dataclass(repr=False)
 class SourcedPublication(Publication):
-    key: str
     sources: list = field(default_factory=list)
 
+    @property
+    def key(self):
+        if self.sources:
+            return self.sources[0].key
+        else:
+            return None
+
     @classmethod
     def from_sources(cls, sources):
-        sources = sorted(sources, key=cls.score_source, reverse=True)
+        sources = sort_publication_sources(sources)
         main = sources[0]
         res = cls(
-            **{k: getattr(main, k) for k in main.__dict__ if k in cls.__match_args__},
+            **{
+                k: getattr(main, k)
+                for k in ["title", "authors", "venue", "type", "year"]
+            },
             sources=sources,
         )
-        for k, v in main.__dict__.items():
-            if k not in cls.__match_args__:
-                setattr(res, k, v)
         return res
 
-    @staticmethod
-    def score_source(source):
-        scores = [v.get(getattr(source, k, None), 0) for k, v in score_rosetta.items()]
-        scores.append(source.year)
-        return tuple(scores)
-
 
 def regroup_authors(auth_dict, pub_dict):
     """
@@ -100,7 +135,7 @@ def regroup_authors(auth_dict, pub_dict):
         pub.authors = [redirection.get(a.key, a) for a in pub.authors]
 
 
-def regroup_publications(pub_dict, threshold=90, length_impact=0.2):
+def regroup_publications(pub_dict, threshold=85, length_impact=0.05, n_range=5):
     """
     Puts together copies of the same publication.
 
@@ -119,17 +154,19 @@ def regroup_publications(pub_dict, threshold=90, length_impact=0.2):
         Unified publications.
     """
     pub_list = [p for p in pub_dict.values()]
+    res = dict()
 
-    p = Process(length_impact=length_impact)
-    p.fit([paper.title for paper in pub_list])
+    if pub_list:
 
-    res = dict()
-    done = np.zeros(len(pub_list), dtype=bool)
-    for i, paper in enumerate(pub_list):
-        if done[i]:
-            continue
-        locs = np.where(p.transform([paper.title])[0, :] > threshold)[0]
-        pub = SourcedPublication.from_sources([pub_list[i] for i in locs])
-        res[pub.key] = pub
-        done[locs] = True
+        p = Process(length_impact=length_impact, n_range=n_range)
+        p.fit([paper.title for paper in pub_list])
+
+        done = np.zeros(len(pub_list), dtype=bool)
+        for i, paper in enumerate(pub_list):
+            if done[i]:
+                continue
+            locs = np.where(p.transform([paper.title])[0, :] > threshold)[0]
+            pub = SourcedPublication.from_sources([pub_list[i] for i in locs])
+            res[pub.key] = pub
+            done[locs] = True
     return res