Merge pull request #8 from fernandogelin/master

Change functions to use HTTP.jl package instead of deprecated Requests and HttpCommon.

REQUIRE

@@ -1,6 +1,5 @@
 julia 0.6
 EzXML 0.5
 JSON
-Requests
-Gumbo 0.4.0 
-HttpCommon 0.3.0
+Gumbo 0.4.0
+HTTP

src/eutils.jl

@@ -44,7 +44,7 @@ export
 
 import EzXML
 import JSON
-import Requests
+import HTTP
 
 const baseURL = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/"
 
@@ -61,7 +61,7 @@ Parameters: db, version, retmode.
 """
 function einfo(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.get(string(baseURL, "einfo.fcgi"), query=params)
+    return HTTP.request("GET", string(baseURL, "einfo.fcgi"), query=params)
 end
 
 """
@@ -74,7 +74,7 @@ retmode, sort, field, datetype, reldate, mindate, maxdate.
 """
 function esearch(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    res = Requests.get(string(baseURL, "esearch.fcgi"), query=params)
+    res = HTTP.request("GET", string(baseURL, "esearch.fcgi"), query=params)
     if get(params, :usehistory, "") == "y"
         set_context!(ctx, res)
     end
@@ -90,7 +90,8 @@ Parameters: db, id, WebEnv.
 """
 function epost(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    res = Requests.post(string(baseURL, "epost.fcgi"), data=params)
+    body = HTTP.escapeuri(params)
+    res = HTTP.request("POST", string(baseURL, "epost.fcgi"), query=body)
     set_context!(ctx, res)
     return res
 end
@@ -104,7 +105,8 @@ Parameters: db, id, query_key, WebEnv, retstart, retmax, retmode, version.
 """
 function esummary(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.post(string(baseURL, "esummary.fcgi"), data=params)
+    body = HTTP.escapeuri(params)
+    return HTTP.request("POST", string(baseURL, "esummary.fcgi"), query=body)
 end
 
 """
@@ -117,7 +119,8 @@ strand, seq_start, seq_stop, complexity.
 """
 function efetch(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.post(string(baseURL, "efetch.fcgi"), data=params)
+    body = HTTP.escapeuri(params)
+    return HTTP.request("POST", string(baseURL, "efetch.fcgi"), query=body)
 end
 
 """
@@ -130,7 +133,8 @@ datetype, reldate, mindate, maxdate.
 """
 function elink(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.post(string(baseURL, "elink.fcgi"), data=params)
+    body = HTTP.escapeuri(params)
+    return HTTP.request("POST", string(baseURL, "elink.fcgi"), query=body)
 end
 
 """
@@ -142,7 +146,7 @@ Parameters: term.
 """
 function egquery(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.get(string(baseURL, "egquery.fcgi"), query=params)
+    return HTTP.request("GET", string(baseURL, "egquery.fcgi"), query=params)
 end
 
 """
@@ -154,7 +158,7 @@ Parameters: db, term.
 """
 function espell(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.get(string(baseURL, "espell.fcgi"), query=params)
+    return HTTP.request("GET", string(baseURL, "espell.fcgi"), query=params)
 end
 
 """
@@ -166,7 +170,7 @@ Parameters: db, rettype, bdata.
 """
 function ecitmatch(ctx::Associative=empty_context(); params...)
     params = process_parameters(params, ctx)
-    return Requests.get(string(baseURL, "ecitmatch.cgi"), query=params)
+    return HTTP.request("GET", string(baseURL, "ecitmatch.cgi"), query=params)
 end
 
 # Create an empty context.
@@ -181,8 +185,9 @@ function set_context!(ctx, res)
     end
 
     # extract WebEnv and query_key from the response
-    contenttype = res.headers["Content-Type"]
-    data = String(res.data)
+    contenttype = Dict(res.headers)["Content-Type"]
+    data = String(res.body)
+
     if startswith(contenttype, "text/xml")
         doc = EzXML.parsexml(data)
         ctx[:WebEnv] = EzXML.nodecontent(findfirst(doc, "//WebEnv"))

src/umls.jl

@@ -31,8 +31,8 @@ export get_tgt,
        get_semantic_types
 
 using Gumbo
-using Requests
-using HttpCommon
+import HTTP
+import JSON
 
 #------------- Endpoints -----------------------
 const uri = "https://utslogin.nlm.nih.gov"
@@ -53,7 +53,7 @@ end
 """
     time_to_last_save(file)
 
-Get how many hours since last save 
+Get how many hours since last save
 """
 function time_to_last_save(file)
     #unix time is GMT
@@ -86,14 +86,14 @@ end
 """
     get_tgt(; force_new::Bool = false, kwargs...)
 
-Retrieve a ticket granting ticket (TGT) using 
+Retrieve a ticket granting ticket (TGT) using
 
 1. UTS username and password OR
 2. apikey
 
 A tgt is valid for 8 hours. Therefore, look for UTS_TGT.txt in the local
-directory to see if it has been recently stored. One can force getting a 
-new ticket by passing keyword argument `force_new=true` 
+directory to see if it has been recently stored. One can force getting a
+new ticket by passing keyword argument `force_new=true`
 
 ####Examples
 
@@ -107,6 +107,7 @@ tgt = get_tgt(apikey = "mykey")
 """
 function get_tgt(; force_new::Bool = false, kwargs...)
     params = Dict(kwargs)
+    body = HTTP.escapeuri(params)
     auth_endpoint = "/cas/v1/tickets/"
 
     if !haskey(params, :apikey)
@@ -118,17 +119,17 @@ function get_tgt(; force_new::Bool = false, kwargs...)
     end
 
     # Check if there is a valid ticket on disk
-    tgt_file = "UTS_TGT.txt"    
+    tgt_file = "UTS_TGT.txt"
     if tgt_exists(tgt_file=tgt_file) && !force_new
-        info("UTS: Reading TGT from file")        
+        info("UTS: Reading TGT from file")
         return readline(tgt_file)
     end
 
     info("UTS: Requesting new TGT")
     headers = Dict("Content-type"=> "application/x-www-form-urlencoded",
     "Accept"=> "text/plain", "User-Agent"=>"julia" )
-    r = post(uri*auth_endpoint,data=params,headers=headers)
-    ascii_r = String(r.data)
+    r = HTTP.request("POST", uri*auth_endpoint, body=body, headers=headers)
+    ascii_r = String(r.body)
 
     doc = parsehtml(ascii_r)
     #for now - harcoded
@@ -154,15 +155,17 @@ Retrieve a single-use Service Ticket using TGT
 """
 function get_ticket(tgt)
     params = Dict("service"=> service)
-    h = Dict("Content-type"=> "application/x-www-form-urlencoded",
+    body = HTTP.escapeuri(params)
+
+    headers = Dict("Content-type"=> "application/x-www-form-urlencoded",
     "Accept"=> "text/plain", "User-Agent"=>"JuliaBioServices" )
-    r = HttpCommon.Response(503)
+    r = HTTP.Response(503)
     try
-        r = post(tgt; data=params, headers=h)
+        r = HTTP.request("POST", tgt; body=body, headers=headers)
     catch
         isdefined(r, :code) ? error("UMLS GET error: ", r.code) : error("UMLS COULD NOT GET")
     end
-    return String(r.data)
+    return String(r.body)
 end
 
 #------------------Search ---------------------
@@ -224,13 +227,13 @@ function search_umls(tgt, query; version::String="current", timeout=1)
         query["ticket"]= ticket
         query["pageNumber"]= string(page)
 
-        r = get(rest_uri*content_endpoint, query=query, timeout=timeout)
+        r = HTTP.request("GET", rest_uri*content_endpoint, query=query, timeout=timeout)
 
         if r.status != 200
             error("Bad HTTP status $(r.status)")
         end
 
-        json_response = Requests.json(r)
+        json_response = JSON.parse(String(r.body))
         # println("No Results ", length(json_response["result"]["results"]))
         if json_response["result"]["results"][1]["ui"]=="NONE"
             break
@@ -260,7 +263,7 @@ end
 """
     get_cui(tgt,cui)
 
-Retrieve information (name, semantic types, number of atoms, etc) for a known CUI 
+Retrieve information (name, semantic types, number of atoms, etc) for a known CUI
 from latest UMLS version or a specific release.
 
 Returns UTS json response
@@ -285,14 +288,14 @@ function get_cui(tgt,cui; version="current")
         rethrow(err)
     end
 
-    r = HttpCommon.Response(503)
+    r = HTTP.Response(503)
     try
-        r = get( rest_uri*content_endpoint,query=Dict("ticket"=> ticket))
+        r = HTTP.request("GET", rest_uri*content_endpoint,query=Dict("ticket"=> ticket))
     catch
         isdefined(r, :code) ? error("UMLS GET error: ", r.code) : error("UMLS COULD NOT GET")
     end
 
-    return Requests.json(r)
+    return JSON.parse(String(r.body))
 end
 
 

test/eutils.jl

@@ -3,23 +3,27 @@
     @testset "einfo" begin
         res = einfo(db="pubmed")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
+        @test nodename(elements(root(parsexml(res.body)))[1]) != "ERROR"
     end
 
     @testset "esearch" begin
-        res = esearch(db="pubmed", term="asthma")
+        res = esearch(db="pubmed", term="""(Asthma[MeSH Major Topic]) AND
+                                        ("1/1/2018"[Date - Publication] :
+                                        "3000"[Date - Publication])""")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
+        @test nodename(elements(root(parsexml(res.body)))[1]) != "ERROR"
     end
 
     @testset "epost" begin
         ctx = Dict()
         res = epost(ctx, db="protein", id="NP_005537")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
         @test haskey(ctx, :WebEnv)
         @test haskey(ctx, :query_key)
     end
@@ -28,13 +32,15 @@
         # esummary doesn't seem to support accession numbers
         res = esummary(db="protein", id="15718680,157427902,119703751")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
+        @test nodename(elements(root(parsexml(res.body)))[1]) != "ERROR"
 
         res = esummary(db="protein", id=["15718680", "157427902", "119703751"])
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
+        @test nodename(elements(root(parsexml(res.body)))[1]) != "ERROR"
 
         # esearch then esummary
         query = "asthma[mesh] AND leukotrienes[mesh] AND 2009[pdat]"
@@ -54,8 +60,8 @@
     @testset "efetch" begin
         res = efetch(db="nuccore", id="NM_001178.5", retmode="xml", idtype="acc")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
 
         # epost then efetch
         ctx = Dict()
@@ -68,22 +74,28 @@
     @testset "elink" begin
         res = elink(dbfrom="protein", db="gene", id="NM_001178.5")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
     end
 
     @testset "egquery" begin
-        res = egquery(term="asthma")
+        res = egquery(term="""(Asthma[MeSH Major Topic]) AND
+                              ("1/1/2018"[Date - Publication] :
+                              "3000"[Date - Publication])""")
         @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
     end
 
     @testset "espell" begin
-        res = espell(db="pubmed", term="athma")
-        @test res.status == 200
-        @test startswith(res.headers["Content-Type"], "text/xml")
-        @test isa(parsexml(res.data), EzXML.Document)
+        res = espell(db="pmc", term="fiberblast cell grwth")
+        @test res.status == 200
+        @test startswith(Dict(res.headers)["Content-Type"], "text/xml")
+        @test isa(parsexml(res.body), EzXML.Document)
+        doc = parsexml(res.body)
+        spelled_query = elements(root(doc))[4]
+        replaced_spell = nodecontent(elements(spelled_query)[4])
+        @test replaced_spell == "growth"
     end
 
     @testset "ecitmatch" begin
@@ -96,4 +108,4 @@
         # requires retmode="xml".
         # @show res.headers["Content-Type"]
     end
-end
+end

test/umls.jl

@@ -1,10 +1,10 @@
 
 
 @testset "UMLS" begin
-    user= ""
+    user = ""
     psswd = ""
     apikey = ""
-    
+
     try
         user = ENV["UMLS_USER"]
         psswd = ENV["UMLS_PSSWD"]
@@ -17,10 +17,9 @@
         term = "obesity"
         query = Dict("string"=>term, "searchType"=>"exact" )
 
-
         tgt = get_tgt(force_new=true, username=user, password=psswd)
         #do it again to test reading from file
-        tgt = get_tgt(username=user, password=psswd)   
+        tgt = get_tgt(username=user, password=psswd)
         #test apikey
         tgt = get_tgt(force_new=true, apikey=apikey)
 
@@ -33,4 +32,4 @@
             @test sm[1] == "Disease or Syndrome"
         end
     end
-end
+end