npcsh 1.1.17__py3-none-any.whl → 1.1.19__py3-none-any.whl

npcsh/npc_team/jinxs/lib/research/paper_search.jinx

@@ -1,101 +1,412 @@
 jinx_name: paper_search
-description: Search for academic papers across multiple sources (Semantic Scholar, arXiv, local datasets)
+description: Search for academic papers with interactive TUI (Semantic Scholar + arXiv)
 inputs:
 - query: ""
-- limit: 10
+- limit: "20"
 - source: "all"
+- text: "false"
+
 steps:
   - name: search_papers
     engine: python
     code: |
       import os
+      import sys
+      import tty
+      import termios
       import time
       import requests
       import urllib.request
       import urllib.parse
       import xml.etree.ElementTree as ET
+      import webbrowser
+      import subprocess
 
       query = context.get('query', '')
-      limit = int(context.get('limit', 10))
+      limit = int(context.get('limit', 20))
       source = context.get('source', 'all').lower()
+      text_mode = context.get('text', '').lower() in ('true', '1', 'yes')
 
       if not query:
-          context['output'] = """Usage: /paper_search <query> [--limit N] [--source SOURCE]
-
-      Sources:
-        all    - Search all available sources (default)
-        s2     - Semantic Scholar only (requires S2_API_KEY)
-        arxiv  - arXiv only
-      """
-          exit()
+          lines = [
+              "Usage: /paper_search <query>",
+              "",
+              "Options:",
+              "  limit  - Max results per source (default 20)",
+              "  source - all (default), s2, arxiv",
+              "  text   - Text-only output, no TUI (true/false)",
+              "",
+              "TUI Controls:",
+              "  j/k or arrows - Navigate",
+              "  1/2/3         - Sort by year/citations/author",
+              "  s             - Filter by source (all/S2/arXiv)",
+              "  y             - Filter by year",
+              "  p             - Preview abstract",
+              "  o             - Open in browser",
+              "  i             - Open in incognide",
+              "  d             - Download paper info",
+              "  q/ESC         - Quit",
+              "",
+              "Examples:",
+              "  /paper_search machine learning",
+              "  /paper_search neural networks source=arxiv",
+          ]
+          context['output'] = "\n".join(lines)
+      else:
+          all_results = []
 
-      all_results = []
+          # Semantic Scholar
+          if source in ['all', 's2']:
+              api_key = os.environ.get('S2_API_KEY')
+              if api_key:
+                  try:
+                      url = "https://api.semanticscholar.org/graph/v1/paper/search"
+                      headers = {"x-api-key": api_key}
+                      params = {"query": query, "limit": limit, "fields": "title,abstract,authors,year,citationCount,url,venue"}
+                      response = requests.get(url, headers=headers, params=params, timeout=30)
+                      response.raise_for_status()
+                      for paper in response.json().get('data', []):
+                          all_results.append({
+                              'source': 'S2',
+                              'title': paper.get('title', ''),
+                              'year': paper.get('year'),
+                              'citations': paper.get('citationCount', 0),
+                              'authors': [a.get('name', '') for a in paper.get('authors', [])],
+                              'abstract': paper.get('abstract', '') or '',
+                              'url': paper.get('url', ''),
+                              'venue': paper.get('venue', '')
+                          })
+                  except Exception as e:
+                      pass
 
-      # Semantic Scholar
-      if source in ['all', 's2']:
-          api_key = os.environ.get('S2_API_KEY')
-          if api_key:
+          # arXiv
+          if source in ['all', 'arxiv']:
               try:
-                  url = "https://api.semanticscholar.org/graph/v1/paper/search"
-                  headers = {"x-api-key": api_key}
-                  params = {"query": query, "limit": limit, "fields": "title,abstract,authors,year,citationCount,url"}
-                  response = requests.get(url, headers=headers, params=params, timeout=30)
-                  response.raise_for_status()
-                  for paper in response.json().get('data', []):
+                  base_url = "http://export.arxiv.org/api/query"
+                  params = {"search_query": f"all:{query}", "max_results": limit}
+                  arxiv_url = f"{base_url}?{urllib.parse.urlencode(params)}"
+                  with urllib.request.urlopen(arxiv_url, timeout=30) as response:
+                      data = response.read().decode('utf-8')
+                  root = ET.fromstring(data)
+                  ns = {'atom': 'http://www.w3.org/2005/Atom'}
+                  for entry in root.findall('atom:entry', ns):
                       all_results.append({
-                          'source': 'Semantic Scholar',
-                          'title': paper.get('title', ''),
-                          'year': paper.get('year'),
-                          'citations': paper.get('citationCount', 0),
-                          'authors': [a.get('name', '') for a in paper.get('authors', [])],
-                          'abstract': paper.get('abstract', '')[:300] if paper.get('abstract') else '',
-                          'url': paper.get('url', '')
+                          'source': 'arXiv',
+                          'title': entry.find('atom:title', ns).text.strip().replace('\n', ' '),
+                          'year': int(entry.find('atom:published', ns).text[:4]),
+                          'citations': None,
+                          'authors': [a.find('atom:name', ns).text for a in entry.findall('atom:author', ns)],
+                          'abstract': entry.find('atom:summary', ns).text.strip(),
+                          'url': entry.find('atom:id', ns).text,
+                          'venue': 'arXiv'
                       })
               except Exception as e:
-                  print(f"S2 error: {e}")
-
-      # arXiv
-      if source in ['all', 'arxiv']:
-          try:
-              base_url = "http://export.arxiv.org/api/query"
-              params = {"search_query": f"all:{query}", "max_results": limit}
-              url = f"{base_url}?{urllib.parse.urlencode(params)}"
-              with urllib.request.urlopen(url, timeout=30) as response:
-                  data = response.read().decode('utf-8')
-              root = ET.fromstring(data)
-              ns = {'atom': 'http://www.w3.org/2005/Atom'}
-              for entry in root.findall('atom:entry', ns):
-                  all_results.append({
-                      'source': 'arXiv',
-                      'title': entry.find('atom:title', ns).text.strip().replace('\n', ' '),
-                      'year': entry.find('atom:published', ns).text[:4],
-                      'citations': None,
-                      'authors': [a.find('atom:name', ns).text for a in entry.findall('atom:author', ns)],
-                      'abstract': entry.find('atom:summary', ns).text.strip()[:300],
-                      'url': entry.find('atom:id', ns).text
-                  })
-          except Exception as e:
-              print(f"arXiv error: {e}")
-
-      if not all_results:
-          context['output'] = f"No papers found for: {query}"
-          exit()
-
-      # Format output
-      results = []
-      for i, paper in enumerate(all_results[:limit], 1):
-          authors = ', '.join(paper['authors'][:3])
-          if len(paper['authors']) > 3:
-              authors += ' et al.'
-          year = paper.get('year', '?')
-          citations = f", {paper['citations']} citations" if paper.get('citations') else ""
-
-          results.append(f"{i}. [{paper['source']}] {paper['title']} ({year}{citations})")
-          results.append(f"   Authors: {authors}")
-          if paper['abstract']:
-              results.append(f"   Abstract: {paper['abstract']}...")
-          results.append(f"   URL: {paper['url']}")
-          results.append("")
-
-      context['output'] = f"Found {len(all_results)} papers:\n\n" + "\n".join(results)
-      context['papers'] = all_results
+                  pass
+
+          if not all_results:
+              context['output'] = f"No papers found for: {query}"
+          elif text_mode:
+              # Text-only output
+              results = []
+              for i, paper in enumerate(all_results[:limit], 1):
+                  authors = ', '.join(paper['authors'][:3])
+                  if len(paper['authors']) > 3:
+                      authors += ' et al.'
+                  year = paper.get('year', '?')
+                  citations = f", {paper['citations']} citations" if paper.get('citations') else ""
+
+                  results.append(f"{i}. [{paper['source']}] {paper['title']} ({year}{citations})")
+                  results.append(f"   Authors: {authors}")
+                  if paper['abstract']:
+                      results.append(f"   Abstract: {paper['abstract'][:200]}...")
+                  results.append(f"   URL: {paper['url']}")
+                  results.append("")
+
+              context['output'] = f"Found {len(all_results)} papers:\n\n" + "\n".join(results)
+              context['papers'] = all_results
+          else:
+              # Interactive TUI mode
+              def get_terminal_size():
+                  try:
+                      size = os.get_terminal_size()
+                      return size.columns, size.lines
+                  except:
+                      return 80, 24
+
+              width, height = get_terminal_size()
+              selected = 0
+              scroll = 0
+              list_height = height - 5
+              mode = 'list'
+              preview_scroll = 0
+              sort_mode = 'relevance'
+              source_filter = 'all'
+              year_filter = None
+
+              def sort_papers(papers, sort_mode):
+                  if sort_mode == 'year':
+                      return sorted(papers, key=lambda x: x.get('year') or 0, reverse=True)
+                  elif sort_mode == 'citations':
+                      return sorted(papers, key=lambda x: x.get('citations') or 0, reverse=True)
+                  elif sort_mode == 'author':
+                      return sorted(papers, key=lambda x: (x.get('authors', [''])[0] if x.get('authors') else ''))
+                  return papers
+
+              def filter_papers(papers, source_filter, year_filter):
+                  result = papers
+                  if source_filter != 'all':
+                      result = [p for p in result if p.get('source', '').lower() == source_filter.lower()]
+                  if year_filter:
+                      result = [p for p in result if p.get('year') and p.get('year') >= year_filter]
+                  return result
+
+              display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+
+              fd = sys.stdin.fileno()
+              old_settings = termios.tcgetattr(fd)
+
+              try:
+                  tty.setcbreak(fd)
+                  sys.stdout.write('\033[?25l')
+                  sys.stdout.write('\033[2J\033[H')
+
+                  while True:
+                      width, height = get_terminal_size()
+                      list_height = height - 5
+
+                      if mode == 'list':
+                          if selected < scroll:
+                              scroll = selected
+                          elif selected >= scroll + list_height:
+                              scroll = selected - list_height + 1
+
+                      sys.stdout.write('\033[H')
+
+                      # Header
+                      if mode == 'list':
+                          sort_ind = {'relevance': '0', 'year': '1', 'citations': '2', 'author': '3'}[sort_mode]
+                          yr = f" year>={year_filter}" if year_filter else ""
+                          header = f" PAPERS ({len(display_papers)}): '{query}' [sort:{sort_mode}({sort_ind}) src:{source_filter}{yr}] "
+                      else:
+                          header = f" PREVIEW: {display_papers[selected].get('title', '')[:width-12]} "
+                      sys.stdout.write(f'\033[44;37;1m{header.ljust(width)}\033[0m\n')
+
+                      # Column headers
+                      if mode == 'list':
+                          col_header = f' {"SRC":<6} {"YEAR":<5} {"CITE":<5} {"AUTHORS":<22} {"TITLE":<50}'
+                          sys.stdout.write(f'\033[90m{col_header[:width]}\033[0m\n')
+                      else:
+                          sys.stdout.write(f'\033[90m{"─" * width}\033[0m\n')
+
+                      if mode == 'list':
+                          for i in range(list_height):
+                              idx = scroll + i
+                              sys.stdout.write(f'\033[{3+i};1H\033[K')
+                              if idx >= len(display_papers):
+                                  continue
+
+                              p = display_papers[idx]
+                              src = p.get('source', '?')[:6]
+                              year = str(p.get('year') or '?')[:5]
+                              citations = str(p.get('citations') or '-')[:5]
+                              authors = ', '.join([a[:12] for a in p.get('authors', [])[:2]])
+                              if len(p.get('authors', [])) > 2:
+                                  authors += '..'
+                              authors = authors[:22]
+                              title = (p.get('title') or '')[:55].replace('\n', ' ')
+
+                              # Color by source
+                              if p.get('source') == 'S2':
+                                  src_color = '\033[34m'  # blue
+                              else:
+                                  src_color = '\033[33m'  # orange/yellow
+
+                              line = f" {src_color}{src:<6}\033[0m {year:<5} {citations:<5} {authors:<22} {title}"
+                              line = line[:width+10]
+
+                              if idx == selected:
+                                  sys.stdout.write(f'\033[47;30;1m>{line}\033[0m')
+                              else:
+                                  sys.stdout.write(f' {line}')
+
+                          # Status bar
+                          sys.stdout.write(f'\033[{height-2};1H\033[K\033[90m{"─" * width}\033[0m')
+                          sel = display_papers[selected] if display_papers else {}
+                          venue = (sel.get('venue') or '')[:40]
+                          sys.stdout.write(f'\033[{height-1};1H\033[K {venue}'.ljust(width))
+                          sys.stdout.write(f'\033[{height};1H\033[K\033[44;37m j/k:Nav 0/1/2/3:Sort s:Source y:Year p:Preview o:Open d:DL q:Quit [{selected+1}/{len(display_papers)}] \033[0m')
+
+                      else:  # preview mode
+                          sel = display_papers[selected]
+                          abstract = sel.get('abstract') or 'No abstract available'
+
+                          preview_lines = [
+                              f"Title: {sel.get('title', '')}",
+                              "",
+                              f"Source: {sel.get('source', '')}",
+                              f"Year: {sel.get('year', '?')}",
+                              f"Citations: {sel.get('citations') or 'N/A'}",
+                              f"Venue: {sel.get('venue', '')}",
+                              "",
+                              "Authors:",
+                          ]
+                          for a in sel.get('authors', []):
+                              preview_lines.append(f"  - {a}")
+                          preview_lines.append("")
+                          preview_lines.append("Abstract:")
+
+                          # Word wrap abstract
+                          words = abstract.split()
+                          line = ""
+                          for w in words:
+                              if len(line) + len(w) + 1 > width - 4:
+                                  preview_lines.append(f"  {line}")
+                                  line = w
+                              else:
+                                  line = f"{line} {w}" if line else w
+                          if line:
+                              preview_lines.append(f"  {line}")
+
+                          preview_lines.append("")
+                          preview_lines.append(f"URL: {sel.get('url', '')}")
+
+                          for i in range(list_height):
+                              idx = preview_scroll + i
+                              sys.stdout.write(f'\033[{3+i};1H\033[K')
+                              if idx < len(preview_lines):
+                                  sys.stdout.write(preview_lines[idx][:width-1])
+
+                          sys.stdout.write(f'\033[{height-2};1H\033[K\033[90m{"─" * width}\033[0m')
+                          sys.stdout.write(f'\033[{height-1};1H\033[K [{preview_scroll+1}/{len(preview_lines)} lines]')
+                          sys.stdout.write(f'\033[{height};1H\033[K\033[44;37m j/k:Scroll b:Back o:Open d:Download q:Quit \033[0m')
+
+                      sys.stdout.flush()
+
+                      c = sys.stdin.read(1)
+
+                      if c == '\x1b':
+                          c2 = sys.stdin.read(1)
+                          if c2 == '[':
+                              c3 = sys.stdin.read(1)
+                              if c3 == 'A':
+                                  if mode == 'list' and selected > 0:
+                                      selected -= 1
+                                  elif mode == 'preview' and preview_scroll > 0:
+                                      preview_scroll -= 1
+                              elif c3 == 'B':
+                                  if mode == 'list' and selected < len(display_papers) - 1:
+                                      selected += 1
+                                  elif mode == 'preview' and preview_scroll < 100:
+                                      preview_scroll += 1
+                          else:
+                              if mode == 'preview':
+                                  mode = 'list'
+                                  sys.stdout.write('\033[2J\033[H')
+                              else:
+                                  context['output'] = "Cancelled."
+                                  break
+                          continue
+
+                      if c == 'q' or c == '\x03':
+                          context['output'] = "Cancelled."
+                          break
+                      elif c == 'k':
+                          if mode == 'list' and selected > 0:
+                              selected -= 1
+                          elif mode == 'preview' and preview_scroll > 0:
+                              preview_scroll -= 1
+                      elif c == 'j':
+                          if mode == 'list' and selected < len(display_papers) - 1:
+                              selected += 1
+                          elif mode == 'preview' and preview_scroll < 100:
+                              preview_scroll += 1
+                      elif c == '0':
+                          sort_mode = 'relevance'
+                          display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+                          selected = 0
+                          scroll = 0
+                      elif c == '1':
+                          sort_mode = 'year'
+                          display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+                          selected = 0
+                          scroll = 0
+                      elif c == '2':
+                          sort_mode = 'citations'
+                          display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+                          selected = 0
+                          scroll = 0
+                      elif c == '3':
+                          sort_mode = 'author'
+                          display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+                          selected = 0
+                          scroll = 0
+                      elif c == 's' and mode == 'list':
+                          # Cycle through source filters
+                          if source_filter == 'all':
+                              source_filter = 's2'
+                          elif source_filter == 's2':
+                              source_filter = 'arxiv'
+                          else:
+                              source_filter = 'all'
+                          display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+                          selected = 0
+                          scroll = 0
+                      elif c == 'y' and mode == 'list':
+                          import datetime
+                          current_year = datetime.datetime.now().year
+                          if year_filter is None:
+                              year_filter = current_year - 1
+                          elif year_filter == current_year - 1:
+                              year_filter = current_year - 3
+                          elif year_filter == current_year - 3:
+                              year_filter = current_year - 5
+                          else:
+                              year_filter = None
+                          display_papers = filter_papers(sort_papers(all_results, sort_mode), source_filter, year_filter)
+                          selected = 0
+                          scroll = 0
+                      elif c == 'p' and mode == 'list' and display_papers:
+                          mode = 'preview'
+                          preview_scroll = 0
+                          sys.stdout.write('\033[2J\033[H')
+                      elif c == 'b' and mode == 'preview':
+                          mode = 'list'
+                          sys.stdout.write('\033[2J\033[H')
+                      elif c == 'o' and display_papers:
+                          paper_url = display_papers[selected].get('url', '')
+                          if paper_url:
+                              webbrowser.open(paper_url)
+                      elif c == 'i' and display_papers:
+                          paper_url = display_papers[selected].get('url', '')
+                          if paper_url:
+                              try:
+                                  subprocess.run(['npcsh', '-c', f'/navigate url={paper_url}'], check=False, capture_output=True)
+                              except:
+                                  webbrowser.open(paper_url)
+                      elif c == 'd' and display_papers:
+                          sel = display_papers[selected]
+                          title = sel.get('title', 'paper')
+                          safe_title = "".join(ch if ch.isalnum() or ch in ' -_' else '_' for ch in title)[:50]
+                          filename = f"{safe_title}.txt"
+                          with open(filename, 'w') as f:
+                              f.write(f"Title: {sel.get('title', '')}\n")
+                              f.write(f"Source: {sel.get('source', '')}\n")
+                              f.write(f"Year: {sel.get('year', '')}\n")
+                              f.write(f"Citations: {sel.get('citations', 'N/A')}\n")
+                              f.write(f"Venue: {sel.get('venue', '')}\n")
+                              f.write(f"Authors: {', '.join(sel.get('authors', []))}\n")
+                              f.write(f"URL: {sel.get('url', '')}\n")
+                              f.write(f"\nAbstract:\n{sel.get('abstract', '')}\n")
+                          context['output'] = f"Saved to: {filename}"
+                          break
+                      elif c in ('\r', '\n') and display_papers:
+                          sel = display_papers[selected]
+                          context['output'] = f"Selected: {sel.get('title', '')}\nURL: {sel.get('url', '')}"
+                          context['selected_paper'] = sel
+                          break
+
+              finally:
+                  termios.tcsetattr(fd, termios.TCSADRAIN, old_settings)
+                  sys.stdout.write('\033[?25h')
+                  sys.stdout.write('\033[2J\033[H')
+                  sys.stdout.flush()