All paths between 2 nodes

Similar Questions

• Calculate the fastest path between 2 nodes with the data model for the Oracle network

  Hi all,
  
  I have Oracle 10 g 2.
  
  My problem is the following:
  
  I created a network named ITALIA_NET in the data model for the Oracle network.
  The table of links of this network is named: ITALIA_NET_LINK$.
  The table of the nodes of this network is named: ITALIA_NET_NODE$.
  
  The table ITALIA_NET_LINK$ contains a field named COST that contains links (in meters) lengths.
  
  I've already calculated the SHORTEST PATH between two nodes of the network, by using the method of shortestPath() (using the Java API) as shown on "Pro Oracle Spatial for Oracle Database 11 g" manual. Infact, this method makes reference to the COST field for $ ITALIA_NET_LINK to make this calculation.
  
  Now, I want to calculate the FASTER PATH between two nodes of the network. I have the links (in hours) travel time to make this calculation.
  
  My idea is to create a new field in ITALIA_NET_LINK$ named Cost2 containing the travel time from the links and then do the math by using the shorthestPath() method, referring to the Cost2 field to $ ITALIA_NET_LINK COST field instead.
  By default, I know that the shorthestPath() method returns the COST field for $ ITALIA_NET_LINK. Is possible to change this setting and do that this method refers to the Cost2 field?
  
  In the alternative, is another way for the calculation of the fastest way?
  I want to leave the creation of another network as last solution, because I will have other costs of field (Cost3, cos4t,...)
  
  Thank you much in advance.

  Your approach is good. You will have two networks and you can read them in memory and analyze of shortest path. The shortestPath method is static for the class of NetworkManager. You can use the same method for both networks, once they are read into memory.

  ...
  read the network with time as cost of link
  NetTime network = NetworkManager.readNetwork (dbConnection, 'ITALIA_NET_TIME');
  read the network length as cost of link
  Network netLength = NetworkManager.readNetwork (dbConnection, 'ITALIA_NET_LENGTH');

  calculate the quickest way
  PathTime path = NetworkManager.shortestPath (netTime, startNodeID, endNodeID);
  calculate the shortest path
  PathLength path = NetworkManager.shortestPath (netLength, startNodeID, endNodeID);
  ...

  In the future, if you upgrade to 11g, network data model provides a load on demand (LOD) API that loads only the scores of necessary network in memory during the analysis. This command removes the restriction of the memory of the 10g (in memory API you use) API. API of LOD can handle very large networks and offers more features analysis and modeling capabilities.

  The following link contains the tutorial of NDM LOD API ready for download. Just for your information.
  https://spatial.SampleCode.Oracle.com/servlets/ProjectProcess?PageID=0Zl7oV

  Kind regards
  Jack

• Find paths between two nodes using SQL sort by cost

  Hi gurus,
  
  I want to find all paths from the source node to the target node that will be sorted by the cost. I don't know if it was possible with the SQL statement. How to start with this query? The script to create the tables and data are given below:
  
  ------------------------------------------------------------------------------------------------------------------
  create table of nodes (nodeId int Primary Key, nodeName varchar (50));
  
  create paths of table (pathId int Primary Key, int fromNodeId, int toNodeId, int cost);
  
  insert into a values (1, 'ISL') nodes;
  insert into a values (2, "LHR") nodes;
  insert into a values (3, 'HYD') nodes;
  insert into a values (4, 'FLS') nodes;
  insert into a values (5, 'MUL') nodes;
  insert into a values (6, "KHI") nodes;
  insert into a values (7, 'QT') nodes;
  
  insert into values of paths (1,1,3,20);
  insert into values of paths (2,1,5,10);
  insert into values of paths (3,1,7,80);
  insert into values of paths (4,2,4,10);
  insert into values of paths (5,3,4,40);
  insert into values of paths (6,3,5,20);
  insert into values of paths (7,3,6,10);
  insert into values of paths (8,6,7,30);
  insert into values of paths (9,6,5,30);
  insert into values of paths (10,6,3,10);
  insert into values of paths (11,7,2,20);
  insert into values of paths (12,5,4,40);
  insert into values of paths (13,5,7,40);
  
  ----------------------------------------------------------------------------------------------------------------------------------
  Assume that the source = ISL and QT= target, there are different paths to QT ISL:
  
  Path relative cost DSB #.
  ==== ================================= =================
  1 ISL-> Multimedia-> QT 50
  2 ISL->->-> QT 60 KHI HYD
  3 ISL-> QT 80
  4 ISL-> HYD-> Multimedia-> QT 80
  5 ISL-> HYD at KHI-> Multimedia-> QT 100
  
  This gives us all possible paths, sorted by price.
  
  No indication or help will be appreciated.
  
  Thanks in advance and best regards
  
  Bilal
  
  Published by: naive2Oracle on February 11, 2011 09:59

  select  ltrim(sys_connect_by_path(nodeName,'->'),'->') path,
          extractvalue(dbms_xmlgen.getXMLtype('select '|| sys_connect_by_path(cost,'+') ||' cost from dual'),'/ROWSET/ROW/COST') - cost cost
    from  paths p,
          nodes n
    where p.fromNodeId = n.nodeId
      and nodeName = 'QT'
    start with nodeName = 'ISL'
    connect by nocycle fromNodeId = prior toNodeId
                   and prior nodeName != 'QT'
      order by cost
  /
  
  PATH                                                              COST
  -------------------------------------------------- -------------------
  ISL->MUL->QT                                                        50
  ISL->HYD->KHI->QT                                                   60
  ISL->HYD->MUL->QT                                                   80
  ISL->QT                                                             80
  ISL->HYD->KHI->HYD->MUL->QT                                        100
  ISL->HYD->KHI->MUL->QT                                             100
  
  6 rows selected.
  
  SQL> 
  

  SY.

• Hierarchical Oracle query help needed - path between the crux of two brothers and sisters

  I want to find the path between two nodes of oracle hierarchical Table.
  
  Consider the following case-
  NodeId - ParentId
  =============
  1 > > > > > > 0
  2 > > > > > > 1
  3 > > > > > > 2
  4 > > > > > > 3
  5 > > > > > > 0
  6 > > > > > > 5
  Here I want to query the database table that if there is a path between nodes 3 and 5?
  The previous query you provided work upwards to the root node.
  
  Here is my expected result, 3-> 2-> 1-> 0-> 5
  
  Yet once if I have a query in the table to get the path between 1 and 3, I want to get out of the way - next
  1-> 2-> 3
  
  Therefore, the query works in both cases. Where ADI root can act as an intermediate or no node.
  
  Can you please guide me how I can get it?
  
  Thank you.

  Hello

  user13276471 wrote:
  I want to find the path between two nodes of oracle hierarchical Table.

  Consider the following case-
  NodeId - ParentId
  =============
  1 >>>>>> 0
  2 >>>>>> 1
  3 >>>>>> 2
  4 >>>>>> 3
  5 >>>>>> 0
  6 >>>>>> 5
  Here I want to query the database table that if there is a path between nodes 3 and 5?
  The previous query

  What application is this? If you're referering to another thread, then post a link, such as {message identifier: = 10769125}

  you provided work upwards to the root node.

  Here is my expected result, 3--> 2--> 1--> 0--> 5

  Yet once if I have a query in the table to get the path between 1 and 3, I want to get out of the way - next
  1--> 2--> 3

  Therefore, the query works in both cases. Where ADI root can act as an intermediate or no node.

  Can you please guide me how I can get it?

  I think you want something like this:

  WITH     bottom_up_from_src    AS
  (
       SELECT     nodeid
       ,     parentid
       FROM     table_x
       START WITH     nodeid      = :src_nodeid
       CONNECT BY     nodeid   = PRIOR parentid
  )
  ,     bottom_up_from_dst     AS
  (
       SELECT     *
       FROM     bottom_up_from_src
      UNION ALL
       SELECT     parentid     AS nodeid
       ,     nodeid          AS parentid
       FROM     table_x
       WHERE     nodeid     NOT IN (
                                        SELECT  nodeid
                                 FROM    bottom_up_from_src
                                    )
       START WITH     nodeid        = :dst_nodeid
       CONNECT BY     nodeid        = PRIOR parentid
  )
  SELECT      :src_nodeid || SYS_CONNECT_BY_PATH (parentid, '-->')     AS display_path
  FROM       bottom_up_from_dst
  WHERE       parentid     = :dst_nodeid
  START WITH  nodeid     = :src_nodeid
  CONNECT BY  nodeid     = PRIOR parentid
  ;
  

  This will show how you can get it from: src_nodeid at dst_nodeid, moving to the top or to the bottom of a hierarchy at a time step. This will work regardless of the fact that
  

  
  
  • : src_nodeid is the ancestor of the: dst_nodeid, or
    
  • : src_nodeid is a descendant of: dst_nodeid, or
    
  • both: src_nodeid and: dst_nodeid are the descendants of another node (e.g. 0).
    

  I hope that answers your question.
  If not, post a small example data (CREATE TABLE and only relevant columns, INSERT statements) and also publish outcomes from these data.
  Explain, using specific examples, how you get these results from these data.
  Always say what version of Oracle you are using (for example, 11.2.0.2.0). It is always important, but particularly so with CONNECT BY queries, because each version since Oracle 7 had significant improvements in this area.
  See the FAQ forum {message identifier: = 9360002}

• Connectivity between two nodes of Virtual Box

  For a few days, I try to establish the network connectivity between two nodes OEL (Oracle Enterprise Linux) in Oracle Virtualbox. I tried everything I have found Google, but still nothing works, everytime I try to ping from one node to another node Hostname/IP is showing unknown host. All the advice everyone here will be great for me.
  
  Published by: 918868 on October 30, 2012 02:30

  When you create a virtual machine, by default VirtualBox allows virtual network card and selects the "Network Address Translation" (NAT) mode for it. In this way the guest can connect to the outside world using the networking of the host and the outside world can connect to services on the feedback that you choose to make visible outside of the virtual machine.

  This default configuration is good probably 95% of users of VirtualBox. VirtualBox is, however, extremely flexible in how it can virtualize networks. It supports multiple virtual NICs per virtual machine, the first four that can be configured in detail in the Manager window. Additional network cards can be configured on the command-line with VBoxManage.

  Source: http://www.virtualbox.org/manual/ch03.html#settings-network

  For more details, please see:

  Virtual networks
  http://www.VirtualBox.org/manual/CH06.html

  Concerning
  Girish Sharma

• How to remove a path between two anchor points

  Hello

  everything is in the title,

  could you tell me please how to remove a path between two anchor points without using the tool Eraser path?

  When I select these two folders and press to remove all disappeared.

  PS: I want to keep the anchor after the removal

  Thank you

  with the Selection tool direct (white pointer), click the path segment (no points) or select slide above without going on something else and tap on delete.

• Regular expression, replace all commas between double quotes ONLY

  Hello

  I want to remove all commas between double quotes:

  I tried under model, but it does not replace all if we have more than one comma in double quotes.

  SELECT REGEXP_REPLACE (123, "45.6" Hello, "John, Mike, Marc", 456, "-Anne, Anna"', ' "([^"]*),([^"]*)" ', '"\1;\2" ', 1, 0) suite

  FROM DUAL;

  Result

  --------------------------------------------------------------------------

  123, '45; 6"," Hello, "John, Mike;" "Marc", 456, "Anne-; Anna.

  the first comma between 'John, Mike; Marc"is not replaced...

  Recursion can do if you insist on the SQL solution

  with

  correct (INP, orig, CNT, res, Step) as

  (select str,

  substr(STR,2),

  -case when substr (str, 1, 1) = ""' then 1 else 0 end,

  substr (STR, 1, 1),

  1

  of (select 123, '45.6,' Hello, ' John, Mike, Marc ", 456, '-Anne, Anna', 'single_value' ' str of union double all the)

  Select "123456789" across double Union

  Select ' ""' of all the double union

  Select "" 123,456,789"" Union double all the

  Choose 123, '45.6,' Hello, ' John, Mike, Marc ", 456, 'Ann-Marie, Anna', 'unique value' ' double str

  )

  Union of all the

  Select the inp

  substr(orig,2),

  CNT + case when substr (orig, 1, 1) = ""' then 1 else 0 end,

  RES | cases where substr (orig, 1, 1) = ','

  so to case when mod(cnt,2) = 1

  then ' ~'

  else «»

  end

  of another substr (orig, 1, 1)

  end,

  Step + 1

  Rectifier

  where orig is not null

  )

  Select the inp, res

  Rectifier

  where the orig is null

  INP	RES
  ""	""
  123456789	123456789
  "123456789".	"123 ~ 456 ~ 789".
  123, '45.6' Hello, ' John, Mike, Marc ", 456, '-Anne, Anna', 'single_value '.	"123," "45 ~ 6 ', Hello," John ~ Mike ~ Marc ", 456," Anne-~ Anna ',' single_value '.
  123, '45.6' Hello, ' John, Mike, Marc ", 456,"Ann-Marie, Anna","unique value ".	"123," "45 ~ 6 ', Hello," John ~ Mike ~ ~ ~ Marc ', 456,' Anne-Marie ~ Anna "," unique value ".

  Concerning

  Etbin

• Retrieve all dates between 2 date

  My dear

  If I have 2 Dates I need sql give any dates between these taches\

  for example

  I have table tab1 in this neck of table called hiredate

  I want to get the lowest dates and max then give all dates between these 2 date

  If the lowest date = 01/01/2014

  and the date max = 20/05/2014

  then sql retive

  01/01/2014

  01/02/2014

  01/03/2014

  .....

  ...

  ..

  TO

  20/05/2014

  Thanks in advance

  Hello

  user222 wrote:

  My dear

  If I have 2 Dates I need sql give any dates between these taches\

  for example

  I have table tab1 in this neck of table called hiredate

  I want to get the lowest dates and max then give all dates between these 2 date

  If the lowest date = 01/01/2014

  and the date max = 20/05/2014

  then sql retive

  01/01/2014

  01/02/2014

  01/03/2014

  .....

  ...

  ..

  TO

  20/05/2014

  Thanks in advance

  If you want rows with the date the lowest, highest and any date between these two then:

  SELECT *- or whatever the columns that you want to

  OF tab1

  WHERE hiredate IS NOT NULL;

  I hope that answers your question.

  If this isn't the case, please post a small example data (CREATE TABLE and only relevant columns, INSERT statements) for all of the tables involved and the results desired from these data.

  In the case of a DML (such as INSERT) operation the sample data should show what look like the paintings before the DML, and results will be the content of the or a modified tables after the DML.

  Explain, using specific examples, how you get these results from these data.

  Always say what version of Oracle you are using (for example, 11.2.0.2.0).

  See the FAQ forum: Re: 2. How can I ask a question on the forums?

• Change all paths in a layer without using a layer appearance attribute

  I suspect I did it wrong for a while, but I don't know how to do it right.

  If I want to change the appearance of all paths in a layer (say, something simple to change their line color), it is pretty easy to target in the layer by clicking on the button of the target layer. However, when I do change (as outline in orange color setting), it appears in the appearance of the layer itself - each element within the layer are not orange line directly in their appearance attributes, they just inherit the appearance of the layer attributes.

  I understand that this is the intended behavior, but I found that it's easier for me to avoid the layer effects and have just each element in a layer to be autonomous and show all its own attributes of appearance without something inheriting from the upper layer (I tend to forget and be confused about when they appear). However, I don't know how to force the appearance of a layer down to its subcomponents (likely the compensation of the appearance of the layer to avoid duplication). Is it some commands I'm missing?

  Now how I do it, it's all in the layer by using the layer target, then UN-target target the highest member of the layer. This allows me to change the appearance of all items in the layer (except the one at the top) without changing the appearance of the layer itself. Then I do the top separately. Seems kludgier than she should be. Any help? Thank you.

  tqrtuomo, I was confused by the present, sometimes as well and made the same thing you describe. I think the key, when selecting on a given layer or group is to click on the 'right' of the 'Target' button in the layers panel to the layer/group level. Make sure that there is no 'circle' around the target button or appearance will be assigned to the layer/group.

• Select all dates between two fields txt. How?

  In a slider form 6i, I need to select all dates between two dates provided. I have two fields text Fld1 and Chp2 with the data type date. There is no table behind the two fields.
  For example
  ': FLD1: = MARCH 1, 2013;
  ': CHP2: = MARCH 10, 2013;
  
  Now I have to choose all dates between 01-mar-2013 and 2013-mar-10 '. How to do?

  Gul says:
  In a slider form 6i, I need to select all dates between two dates provided. I have two fields text Fld1 and Chp2 with the data type date. There is no table behind the two fields.
  For example
  ': FLD1: = MARCH 1, 2013;
  ': CHP2: = MARCH 10, 2013;

  Now I have to choose all dates between 01-mar-2013 and 2013-mar-10 '. How to do?

  As'salamualikum Gul

  Try this

  SELECT TO_DATE('01-MAR-2013', 'DD-MON-RRRR') - 1 + rownum AS d
  FROM ALL_OBJECTS
  WHERE TO_DATE('01-MAR-2013', 'DD-MON-RRRR') - 1 + rownum BETWEEN TO_DATE('01-MAR-2013', 'DD-MON-RRRR') AND TO_DATE('10-MAR-2013', 'DD-MON-RRRR')
  

  Hope this helps

• Make a group without selection of all paths

  I'm working on a script that will take each pathItem in a document and put it in a group (all pathItems in the document are groupless opening, so there won't be other groups already in the document).

  I have a script that creates a group, and then puts the highest pathItem placed in the group. I just can't figure out how to retrieve all the pathitems in the document, or the same action in a loop until everything is in a single group. The kicker is, as the title suggests this script should work without selecting anything first hand (in other words, after click on an item to select). Now, if there was a way to select all elements with a script that would be great.

  Here's what I have so far:

  #target Illustrator
  //based on script from Carlos Canto
  var doc = app.activeDocument
  
  if ( app.documents.length > 0 ) {
  doc = app.activeDocument;
  newGroupItem = doc.activeLayer.groupItems.add();
  }
  
  var gi = doc.groupItems[0]; // the topmost existing group
  var all = doc.pathItems[0]; // all paths in document
  
  all.move (gi, ElementPlacement.PLACEATBEGINNING); // move selected path inside the group
  all.evenodd = true; // necessary to determine "insideness" or to make holes.
  
  

  I also have what I found here on the forum

  #target Illustrator
  // GroupFromSel.jsx
  // regards pixxxelschubser 
  if ( app.documents.length > 0 && app.activeDocument.selection.length > 0) {
      var aDoc = app.activeDocument;
      var Sel = aDoc.selection;
      var GroupFromSel = aDoc.groupItems.add();
      for ( i = Sel.length-1; i >= 0; i--) {
          Sel[i].moveToBeginning( GroupFromSel )
          }
  }
  
  

  But alas, you select so two scripts, two solutions possible. I'm just working on both until I can get one to work.

  Any ideas?

  var all = doc.pathItems[0]; // all paths in document

  It's only the highest pathItem, that to change

  var all = doc.pathItems; // all paths in document
  

  .. .and loop thru 'all' and move each within your group

  #target Illustrator
  //based on script from Carlos Canto
  
  if ( app.documents.length > 0 ) {
      doc = app.activeDocument;
      var gi = doc.activeLayer.groupItems.add();
      var all = doc.pathItems; // all paths in document
  
      for (i=0; i
  for this configuration, we will run trouble if you have other groups with pathItems
  		   

• How to return all dates between two dates?

  Hi all
  
  I need get all dates between two dates, Oracle SQL and PL/SQL.
  
  for example

  01/10/2011 - date 1
  10/11/2011 - date 2
  
  # Return Values
  01/10/2011
  10/02/2011
  03/10/2011
  .
  .
  .
  09/11/2011
  10/11/2011

  How can I get these values? Any help will be useful.
  
  Thank you

  Maybe

  select :start_date + level - 1 the_date
    from dual
  connect by level <= :end_date - :start_date + 1
  

  Concerning

  Etbin

• How to hide all paths without widening the palette?

  It is something that I have wondered about for years:

  You know how you can deselect one way 'sex' (for lack of better word) in the palette drawn by clicking in the empty space of the paths palette? What happens if your pallet traced has saved paths as long as there is no space empty to click in? I know that I can just to expand the palette, but I don't want to have to do that, because all my palettes are solidly organized in a giant wall o ' pallets on a second monitor. Is there a way to hide all paths without having to expand the palette to bring up the empty space?

  If it is not currently using a tool running on trails, activate one, for example, press a key. Now, press ESC to untarget all paths.

• It is possible to select a segment of a path between two points outside their clipping mask?

  I have a way with 3D rotation effect tends in a clipping mask. In the Preview mode (Ctrl + Y), most of the road is not the clipping mask, but most of the visible inside the mask in normal view because of the 3D effect turns. I want to select and remove a segment of the path between two points outside the clipping path, but Illustrator does not select the components of access outside their clipping mask path.

  Anyway is to do without for as much free mask?

  ALWAYS indicate the version you are using. Regulars here using one of the six different versions of Illustrator. Assuming that the current version:

  White pointer: marquee select around a portion of the original route which is located in the clipping mask.

  OR

  Black hand: select the object. Make a right click the object. Select isolate selected in the popup clipping mask.

  JET

• XPath to capture nodes between two nodes

  Hi all

  I met a problem a seemingly simple challenge of mine.

  I have the following xml code.

  Basically, I would like to capture the text between the begin-end blocks

  <? XML version = "1.0" encoding = "UTF-8"? >

  < document >

  < body >

  < paragraph >

  < run >

  < fld type = 'start' / >

  < / run >

  < run >

  < text > SAMPLE < / text >

  < / run >

  < run >

  < text > < / text >

  < / run >

  < run >

  < text > FIELD < / text >

  < / run >

  < run >

  < fld type = 'end' / >

  < / run >

  < run >

  < text > TEXT UNWANTED SOME < / text >

  < / run >

  < / paragraph >

  < paragraph >

  < run >

  < fld type = 'start' / >

  < / run >

  < run >

  < text > SAMPLE < / text >

  < / run >

  < run >

  < text > < / text >

  < / run >

  < run >

  < text > Field2 < / text >

  < / run >

  < run >

  < fld type = 'end' / >

  < / run >

  < run >

  < text > TEXT UNWANTED SOME < / text >

  < / run >

  < run >

  < fld type = 'start' / >

  < / run >

  < run >

  < text > SAMPLE < / text >

  < / run >

  < run >

  < text > < / text >

  < / run >

  < run >

  < text > FIELD3 < / text >

  < / run >

  < run >

  < fld type = 'end' / >

  < / run >

  < / paragraph >

  < / body >

  < / document >

  So far, what I came up with is:

  WITH temp AS (
  SELECT 
         XMLTYPE ('<?xml version="1.0" encoding="UTF-8"?>
                    <document>
                       <body>
                          <paragraph>
                             <run>
                                <fld type="begin" />
                             </run>
                             <run>
                                <text>SAMPLE</text>
                             </run>
                             <run>
                                <text> </text>
                             </run>
                             <run>
                                <text>FIELD</text>
                             </run>
                             <run>
                                <fld type="end" />
                             </run>
                             <run>
                                <text>SOME UNWANTED TEXT</text>
                             </run>
                          </paragraph>
                          <paragraph>
                             <run>
                                <fld type="begin" />
                             </run>
                             <run>
                                <text>SAMPLE</text>
                             </run>
                             <run>
                                <text> </text>
                             </run>
                             <run>
                                <text>FIELD2</text>
                             </run>
                             <run>
                                <fld type="end" />
                             </run>
                             <run>
                                <text>SOME UNWANTED TEXT</text>
                             </run>
                             <run>
                                <fld type="begin" />
                             </run>
                             <run>
                                <text>SAMPLE</text>
                             </run>
                             <run>
                                <text> </text>
                             </run>
                             <run>
                                <text>FIELD3</text>
                             </run>
                             <run>
                                <fld type="end" />
                             </run>
                          </paragraph>
                       </body>
                    </document>') AS xml_template
  FROM dual
  )
  SELECT x.*
    FROM temp
        ,XMLTABLE('//*/paragraph/run[fld/@type="begin"]/following-sibling::run[following-sibling::run[fld/@type="end"]]/text'
                  PASSING temp.xml_template
                  COLUMNS res varchar2(100) PATH '.'
                 )x
  

  Result:

  RES

  SAMPLE

  FIELD

  SAMPLE

  FIELD2

  SOME UNWANTED TEXT

  SAMPLE

  FIELD3

  Unfortunately what I wrote XPath is not good enough, because with regard to the text between the two blocs of start-end it still captures the unwanted text.

  My XPath knowledgege dept unfortunately not so I would appriciate annything you can help me with even an idea could get through on the edge.

  Thank you

  Steve

  Steve,

  XQuery can help in this case:

  SELECT x.*
  FROM temp
     , XMLTABLE('for $begin in /document/body/paragraph/run[fld/@type="begin"]
                 let $end := $begin/following-sibling::run[fld/@type="end"][1]
                 return $begin/following-sibling::run[. << $end]'
         PASSING temp.xml_template
         COLUMNS res varchar2(100) PATH 'text'
       ) x  ;
  

  Explanation:

  This FLWOR expression traverses all nodes 'start' in the document, finds the corresponding node of the 'end' and returns the sequence of "run" nodes follows 'begin' and above 'end '.

  If you want to concatenate all the fragments of text in a block, you can also do it directly, like this:

  for $begin in /document/body/paragraph/run[fld/@type="begin"]
  let $end := $begin/following-sibling::run[fld/@type="end"][1]
  return string-join($begin/following-sibling::run[. << $end]/text, "")
  

  She's going back 3 ranks:

  RES

  -------------

  EXAMPLE OF A FIELD

  SAMPLE FIELD2

  SAMPLE FIELD3